Exploring and Exploiting the Role of Food Structure in Digestion
- 781 Downloads
The structural diversity and complexity encountered in food systems is remarkable in its scope. This is true for both wholefood materials, such as fruit and vegetables, but also for manufactured foods constructed through formulation and process design. The material and structural properties of foods are well known to have a predominant effect on sensory properties and perception, that in turn has a determining role on our food preferences. However, the interactions and assembly of the structural elements in food materials, and the manner in which they are broken down during consumption and digestion, is increasingly seen as of importance in consideration of the nutritional value of what we eat. In fact, a noteworthy aspect of our digestive process is its ability to effectively extract nutritional value from the array of foods available for consumption. However, the response of food structures during digestion can, with appropriate understanding, be manipulated allowing for the enhancement of nutritional value. This chapter explores how the digestive properties of the primary macronutrient components of protein, fat and carbohydrates are influenced by their structural assembly in foods.
KeywordsMicrostructure Digestion Colloids Lipids Protein Carbohydrate
It is well understood that our health and physical well-being is intimately related to what we eat. However, dietary practice at an individual level is a complex issue, being determined by a number of aspects such as personal preference, choice, cultural definition, food availability and affordability, amongst others. To maintain an acceptable quality of life, our daily food intake must comprise an appropriate balance of macronutrients and micronutrients. Whilst the range of food available to us should readily supply such nutritional needs, it needs recognising that our nutritional requirements are actually ultimately met (or not met, as the case may be), based as much on the foods that we would prefer to eat, or perhaps can afford to eat.
It is also increasingly becoming apparent that nutritional content of a food does not necessarily reflect the nutritional value of that same food once consumed. Nutritional labelling on products gives a relative indication as the content of a food in basic terms of protein, carbohydrate and fat, as well as some indication of vitamins and minerals. However, we now have greater understanding that there can be considerable variation in the nutritional value of different protein types, for example in regard to the relative abundance of essential amino acids. Likewise, the nutritional value of lipids can vary according to compositional variation between different fat and oil types, notably in the degree of saturation or unsaturation, and acknowledging that the greatest nutritional value may come from raw materials abundant in polyunsaturated triglycerides. Carbohydrates provide equal complexity in their nutritional representation, in that the term carbohydrate can encompass simple monosaccharide or disaccharide sugars through to oligosaccharides and long chain biopolymers such as starches and polysaccharides. Needless to say that within each of these particular classes of carbohydrates, there can be considerable diversity of materials.
Similar findings were observed regarding the bioavailability of other lipophilic vitamins and micronutrients (carotenoids) through work undertaken by Agustiana, Zhou, Flendrig, and White (2010), which demonstrated negligible uptake of these nutrients when consumed in a salad based meal in the absence of fat. Inclusion of fat in the form of a salad dressing showed improved uptake, with increasing absorption observed as fat levels from the dressing were increased. These studies provide pertinent examples of how the nutritional value of food can be greatly influenced by even small changes to composition and the manner in which it is consumed.
Such studies invariably lead to the question of the broader role of food structure in nutrient digestion and uptake. To a degree this is a relatively recent consideration, possibly due to the fact that whilst our diet comprises a highly diverse array of structured food materials, our digestive system demonstrates a remarkable ability to (almost entirely) utilise nutritional value regardless of the structure of the food in question, and thus its contribution to the digestive process has tended to be ignored. However, it is increasingly being recognised that the structural assembly of nutrients, and the nature of their interactions within a food system—ranging from molecular to material—can be highly impactful on the manner in which those nutrients are digested, and potentially provides a lever by which the health and wellness value of a food, whether wholefood or manufactured, can be optimised. This chapter aims to provide some context regarding the effects of food microstructure on the primary nutrient systems present in foods.
2 Defining Food Structure
The role of the digestion process can be grossly simplified as a mechanism by which ingested nutrients can be rendered into a state amenable for assimilation and utilisation by the body. This simple statement belies the complex biological pathway that has evolved to achieve this effect and which, as indicated, provides humans with the ability to consume, and gain nutritional value, from a vast array of different foodstuffs, and which have equally large diversity in their structural complexity. The study and characterisation of food structure as a scientific discipline has become increasingly prominent over the past few decades, not only in developing a greater understanding as to the relationship between food structure and digestion (Lundin & Golding, 2009; Norton, Wallis, Spyropoulos, Lillford, & Norton, 2014) but also in relation to other attributes of food materials, such as their sensory properties and shelf-life (as determined by physical, chemical and microbiological stability).
Accordingly, attempts have been made to specifically define the concept of food structure. Aguilera summarises two particular definitions, one from 1993, which states that “food microstructure can be defined as the spatial arrangements of elements in a food and their interactions” and a second and similar definition from 1980, determining that structure is “…the organisation of a number of similar of dissimilar elements, their binding into a unit, and the interrelationship between the individual elements and their groups”. This common theme of spatial arrangement and interactions can be applied over multiple length scales from molecular through to microscopic (e.g. colloidal particles, cellular organelles), mesoscopic (e.g. colloidal aggregates, gel structure, cellular assemblies) and material, and with the integration of these length scales defining the overall properties of food.
In this regard, all raw materials contributing to our diet will possess an inherent nature-defined structure which is constructed over time during the farming, harvesting and post-harvest treatment of that material. For some foodstuffs, mainly wholefoods such as fruit and vegetables, it is the naturally assembled food structure that will be extant at point of consumption (whilst acknowledging that gradual changes to structure will inevitably occur post harvesting). For wholefoods commonly consumed raw, this endogenous structure will be determinant on the sensory characteristics of the food, as well as having the potential to influence the subsequent digestion behaviour of the nutrients present within that food. For other wholefoods, such as meat and fish and vegetables, additional processing (usually in the form of heating) is commonly applied prior to consumption. Whilst the primary purpose of such processing has been to ensure that the food is safe for eating, we also recognise the role of heating in imparting desirable sensory properties such as the development of flavours (e.g. through the Maillard reaction) and textural attributes (such as those arising from the denaturation of proteins, breakdown of cellular structures, the melting of fat or gelatinisation of starch). The development of sensory characteristics during what has effectively become the cooking process, is a consequence of changes taking place to the native structure of the food material in response to the applied conditions, noting that such changes will also invariably impact on the digestive properties and nutritional value of the food.
Extending this further, as part of modern food manufacturing, it is also the case that we can use processing pathways to effectively disassemble the native structure of particular raw materials for subsequent use as ingredients in the production of processed foods. The degree of structural modification or disruption can be altered across the various length scales dependent on the requirements of the product. Notably, the production of ingredients with highly defined functional or sensory properties, for example fats and oils derived from plant materials, requires processes capable of complete disruption of the spatial arrangement and interactions of the various structural components within the raw material, thereby enabling the separation, extraction and concentration of the particular molecular component(s) of value. It is important to note that the processing required to achieve separation can potentially impact on the molecular structure of the ingredient being isolated. For relatively non-labile materials that can be separated, refined and concentrated using relatively mild processing conditions (such usually the case for simple sugars, fats and oils), molecular structure is not necessarily altered during the production of the ingredient. However, for more labile components the native molecular structure may be modified by processing, such as for proteins for which denaturation may occur as a consequence of thermal treatments applied during production. Any such modifications at a molecular level will invariably influence the functionality of those ingredients when used in manufacture of a processed food, as well as potentially impacting on the manner in which those materials can be digested.
The definitions of food structure as provided by Aguilera can equally be applied to the production of manufactured or processed foods. In this case, the structural elements are determined by the formulation of the product. Interactions between elements will be dependent on the functionality of the raw materials comprising a formulation (noting that not all ingredients may play a functional role in creation of a food structure, as well as the potential use of additive systems capable of providing highly specific functional roles in food products). However, the enabling of interactions and creation of an appropriate spatial organisation of structural elements is equally achieved through the processing operations utilised in the manufacture of the food (such as thermal treatment, homogenisation and shear). This assembly of structure through formulation and process design determines the material and product properties, enabling manufacturers to produce foods with well-defined and highly reproducible attributes.
The assembly of food structures (whether wholefood or manufactured) towards the point of consumption is invariably a dynamic process (Fundo, Quintas, & Silva, 2015). For manufactured foods, processing during production constitutes the main dynamic pathway for constructing a particular food structure. However, it is important to acknowledge that food structures can continue to be modified through physical, biochemical or chemical means during distribution, storage and utilisation. Likewise, the consumption and digestion of food is a dynamic process, in which food structures are broken down, initially in the mouth and subsequently during gastrointestinal transit.
The deconstruction of the material and structural properties of a food under conditions of oral processing (i.e. the combined role of in mouth shear, combination with saliva, temperature and time) serves to define the sensory characteristics of a food, creating a temporal profile of texture, flavour and taste perception (Foster et al., 2011). Oral processing also serves to render foods in an appropriate structural state for both swallowing and the onset of digestion in the stomach. In this regard, the structural state of liquid foods, such as milk, may be relatively unaltered during the short oral residence time for consumption. In contrast, the structures of solid, or semi-solid foods may be considerably broken up and mixed with saliva in the formation of a swallowable bolus. Accordingly, the digestive behaviours of a bolus formed through oral processing may represent a largely modified structure relative to that of the food system prior to consumption (Bornhorst & Singh, 2012; Wang & Chen, 2017). The following sections will explore the role of hierarchical assembly and food structure on the digestion of the main macronutrient components of protein, fat and carbohydrate.
3 The Influence of Food Structure on Protein Digestion
Mapping of the human proteome has determined that the human body comprises approximately 100,000 different proteins, providing a plethora of highly specific physiological functions. Maintenance and replenishment of the human proteome is supported through consumption and digestion of various sources of dietary protein, which can be broken down initially into oligopeptides before further hydrolysis to free amino acids that are transportable across the small intestinal epithelium. Without expanding into detail regarding the entire physiological processes of protein digestion and uptake, the reduction of dietary protein to smaller peptide fragments occurs through exposure to gastric and pancreatic endopeptidases (pepsin, trypsin and chymotrypsin) with subsequent breakdown into amino acids being achieved through interaction of oligopeptide fragments with digestive exopeptidases. The biochemical hydrolysis of proteins is facilitated by the variable pH conditions in the stomach and small intestine. Thus, the gastric protease pepsin, has optimum activity in the pH range 1.5–2.2 (Schlamowitz & Peterson, 1959), whilst the pancreatic lipases have optimal activity in the range 7.5–8.2.
3.1 Influence of Protein Molecular Structure on Digestion Behaviour
The rate and efficacy by which proteins can be digested can be influenced across all length scales representative of food structure. At the shortest length scale, the interaction between the digestive endopeptidases and the protein substrate can be affected by the molecular conformation of the protein. In the specific case of pepsin or chymotrypsin, which exhibit preferential hydrolysis at peptide bonds comprising hydrophobic or aromatic amino acids (such as phenylalanine or tyrosine), the internalisation of these residues within the native, folded structure of a globular protein can render the protein more resistant to digestion. This has been observed for the milk protein β-lactoglobulin, which has been shown under in vitro gastric conditions as undergoing only limited hydrolysis by pepsin or chymotrypsin whilst in the native state (Loveday, Peram, Singh, Ye, & Jameson, 2014; Mullally, Mehra, & FitzGerald, 1998; Schmidt & Vanmarkwijk, 1993). In contrast, the more disordered molecular structure of the casein protein fraction in milk provides greater exposure of preferential sites for proteolysis and thus (without considering any structural contribution) the protein tends to be far more readily hydrolysed (Guo, Fox, Flynn, & Kindstedt, 1995).
Prior studies on β-lactoglobulin molecular structure (Iametti, DeGregori, Vecchio, & Bonomi, 1996; Townend, Herskovits, & Timasheff, 1969) have determined that significant proportion of amino acids residues susceptible to hydrolysis to be either hidden through quaternary interactions or buried within the protein secondary and tertiary structures, creating a hydrophobic cavity that is not readily accessible by the enzyme. Perhaps understandably, unfolding of the native protein structure and exposure of these buried hydrophobic/aromatic domains has been seen to greatly increase the susceptibility of the protein to not only peptic hydrolysis, but also when exposed to trypsin and chymotrypsin (Iametti et al., 1996). This demonstrates one potential consequence of food processing on protein digestion, in which the denaturation of the protein through various mechanisms leads to an enhancement in the extent of proteolysis such through the application of thermal processing (Kitabatake & Kinekawa, 1998; Mullally et al., 1998). The application of static high pressure represents another processing pathway allowing for enhancement of peptic digestion. This is exemplified in a study by Zeece and co-authors (Zeece, Huppertz, & Kelly, 2008), who demonstrated that pressures in excess of 600 MPa for at least 10 min were able to greatly enhance the digestion of 1 wt% β-lactoglobulin solutions under simulated gastric conditions, observing the complete disappearance of the primary structure occurred after only 1 min of in vitro incubation after pressure treatment. Proteolysis has also been shown to be improved by other mechanisms capable of exposing suitable hydrolysis sites. This includes protein denaturation through addition of urea (Guo et al., 1995), and even the unfolding of the protein as a consequence of adsorption at the oil–water interface when used for emulsification (Macierzanka, Sancho, Mills, Rigby, & Mackie, 2009; Nik, Wright, & Corredig, 2010).
Modification to molecular structure through these various mechanisms can also impact digestion as a consequence of additional intermolecular interaction and self-assembly. Again, this can occur across increasing length scales from mesoscopic to macroscopic, creating structures ranging from oligomeric to colloidal through to percolating gel structures. The extent and nature of aggregation is dependent on a number of variables, including protein concentration, solvent conditions (pH and ionic environment), processing conditions, and their influence on the specific bonding mechanisms responsible for association (i.e. hydrogen bonds, electrostatic interactions, covalent linkages and hydrophobic effects). It can also be noted that association can occur between like molecules, either directly or via an appropriate bridging mechanism, or between different moieties. Whilst the formation of aggregated structures can impact on digestion (again, primarily considering this from the perspective of enzymatic hydrolysis) due to the alterations to the accessibility of peptide linkages susceptible to hydrolysis, other factors can influence digestibility.
3.2 Influence of Intermolecular Interactions on Protein Digestion
Cross-linking mechanisms between proteins and other molecular species has been demonstrated as impacting on the bioavailability of particular amino acids groups that are involved in cross-linking. This is exemplified by the Maillard reaction, which can occur between proteins and reducing sugars, in which the bioavailability of the essential amino acid lysine (along with other reactive amino acids such as arginine, methionine, tryptophan, and histidine) can be compromised as a consequence of cross-linking mechanisms between the amino acid and the sugar (Obrien & Morrissey, 1989). Lysine is reported as supporting a variety of physiological functions including the production of carnitine, lowering cholesterol levels, as well as assisting in the absorption and conservation of calcium. The Maillard reaction not only negates the metabolic availability of lysine, but can also influence the ability of peptidases to hydrolyse particular sequences on the protein chain and leading to the formation of different polypeptide fragments (Schumacher & Kroh, 1996). This can reduce the nutritional value of the reacted protein, and can be problematic for processes and formulations that facilitate the Maillard reaction, such as those occurring during the manufacture and storage of milk powders and sterilized milks, in which the protein component of the milk can form cross-links with lactose during thermalisation (Guyomarc’h, Warin, Muir, & Leaver, 2000; Mehta & Deeth, 2016). For infant nutrition this can be a particular issue, noting that the manufacture, and especially the extended storage, of both UHT and powder based formulations can result in a reduction to biologically available lysine, which is considered a critical amino acid for supporting early growth and development (Ferrer et al., 2003).
The specific loss of amino acid bioavailability as a consequence of Maillard complexation is, however, not necessarily exhibited in other forms of intermolecular cross-linking. For example, biochemical protein polymerisation can be mediated through the use of the enzyme transglutaminase, which catalyses the formation of intermolecular covalent bridges between lysine and glutamine amino acids. Studies on the digestion of transglutaminase cross-linked proteins have demonstrated that exopeptidase digestion is able to cleave the lysine-glutamate isopeptide, liberating free lysine (Romeih & Walker, 2017). That said, the polymerisation of proteins through transglutaminase can influence their susceptibility to hydrolysis via endopeptidase action. For example, in vitro studies on the enzymatic digestion of soy protein isolate carried out by Tang and co-authors (Tang, Li, & Yang, 2006) determined that the rate and extent of protein hydrolysis (as measured by % nitrogen release) was inhibited as a consequence of transglutaminase polymerisation of the protein. This was observed for both proteolysis by pepsin and subsequent digestion by trypsin. Interestingly, Tang observed that whilst heating of the soy protein enhanced peptic digestion by exposing hidden hydrolysis sites through unfolding of the native structure, denaturation was also able to provide greater access to lysine and glutamine sites for transglutaminase polymerisation. Accordingly, the more extensive cross-linking of the heat-treated soy protein was found to result in a marked decrease in peptic digestion (although this was less pronounced for trypsin hydrolysis).
A separate study by Monogioudi and co-authors (Monogioudi et al., 2011) investigated the effects of transglutaminase polymerisation of the milk protein β-casein in relation to its proteolysis by pepsin. They observed a retardation in the rate and reduction in the extent of hydrolysis that increased with extent of protein cross-linking. In vitro studies revealed that after extended incubation, the degree of proteolysis of the non-cross-linked casein was 50% higher than that of the fully cross-linked protein. Interestingly, whilst there was a clear inhibition of proteolysis, the peptides that were generated during digestion were essentially the same, regardless of whether the protein had been cross-linked or not.
Curiously, a prior study undertaken by Roos and co-authors (Roos, Lorenzen, Sick, Schrezenmeir, & Schlimme, 2003) exploring the effect of transglutaminase cross-linking on the gastric and small intestinal hydrolysis of sodium caseinate, the authors determined that the extent and pathway of proteolysis was unaffected by the transglutaminase cross-linking of the protein. Their initial in vitro findings were following up with a mini-pig in vivo study, which demonstrated no significant differences in either the kinetics of digesta flow, or in the total protein digestibility between the cross-linked and untreated protein samples.
Other process-induced intermolecular interactions can have significant consequences for the digestibility of foods without necessarily impacting on the material or sensory properties of those systems. An example of this has been presented by Ye and co-authors (Ye, Cui, Dalgleish, & Singh, 2016b) who studied the effects of heat treatment on the in vitro coagulation and gastric digestion of milk. Their findings showed significant differences in the structure of clot formation, with a milk sample heated at 90 °C for 20 min displaying a much looser clot structure compared to that of an unheated milk. The denser protein network of the unheated milk under gastric pH was found to be more resistant to peptic hydrolysis (as observed for both casein and whey protein digestion), due to limited structural breakdown during gastric incubation, and restricted diffusivity of the enzyme into the interior of the structure. These effects were attributed to the cross-linking of whey proteins to the surface of the casein micelle structure during heating, which significantly altered the interactions and structuring of the casein micelles on lowering of pH towards the isoelectric point. Interestingly, a repeat of the study using whole milk indicated that heat treatment also affected the location and distribution of the fat droplets within the gastric environment, noting that fat droplets were more readily liberated from the heated milk coagulum compared to that of the unheated sample (Ye, Cui, Dalgleish, & Singh, 2016a).
3.3 Influence of Protein Polymers and Self-assemblies on Protein Digestion
The self-assembly of protein molecule into moieties of varying size and structure can be generated through other mechanisms, leading to the formation of a range of structural assemblies such as fibrillar, stranded, branched and random aggregate. These have been termed giant supramolecules, and whilst the assembly of these may extend into colloidal dimensions, the structural dimensions are not yet sufficient to generate fully percolating networks characteristic of protein gels [although these structures may cause weak gelation through entanglement and interaction at sufficiently high volume fractions (Veerman, Sagis, & van der Linden, 2003)]. Formation of these aggregated supramolecules can be achieved through a number of predominantly globular proteins, with whey, egg and soy proteins providing examples of edible proteins known to exhibit this behaviour (Akkermans et al., 2007; Veerman et al., 2003). The mechanisms of assembly of these structures, and the consequences of these on material and functional properties has been an area of considerable research interest (Zhao, Pan, & Lu, 2008), with the digestive properties of these assemblies coming under particular scrutiny in recent years (Moayedzadeh, Madadlou, & Asl, 2015).
Findings from this study have indicated that, whilst there is potential value in the technical functionality of these supramolecular structures, further work is required to ensure that any variance in digestion behaviours observed for such structures does not pose any health risk arising from their consumption.
3.4 Influence of Protein Gels and Protein-Structured Foods on Protein Digestion
Much of the discussion thus far has centred on how protein digestion is influenced through alterations to molecular conformation, either at an individual level or within complexes, self-assembled and polymerised structures. By and large, the material properties of these systems are not considered as providing meaningful contribution to their digestive behaviours, with investigations carried out with the protein essentially present in a fluid state in solution (and usually under relatively dilute conditions). However, aggregation of protein structures (whether in isolation, or in combination with other components) can be extended such that percolating networks are formed, resulting in the transition of material state from liquid to solid and the formation of gelled structures. In addition to changes to molecular conformation caused by the creation of the gelled state that may influence susceptibility to digestion, the change in material state is likely to influence the diffusivity and accessibility of the digestive enzyme in relation to the protein substrate.
A commonly cited example is the difference in digestion of raw and cooked egg protein. From a molecular perspective, the native structure of the globular proteins comprising egg white represent a more resistant conformation for enzymatic hydrolysis, as discussed in Sect. 3.1. This is in spite of the fact that the physical state of raw egg is liquid, and therefore should be expected to provide a more effective medium for enzymatic diffusion into the ingested material relative to the gelled cooked egg (acknowledging that the structure of the cooked egg would have been comminuted to varying degrees through mastication). In contrast, the denaturation and unfolding of the egg protein as consequence of heating would be expected to provide a more amenable substrate for proteolysis. This hypothesis was explored in a human study undertaken by Evenepoel and co-authors (Evenepoel et al., 1998), who used a stable isotope method to determine both the assimilation and gastric emptying time of raw and cooked egg. Raw egg was found to be significantly less digestible over the entire course of digestion compared to the cooked egg. In fact, the total ileal digestibility of the raw egg was determined as 51.3% (±9.8) compared to 90.9% (±0.8) for the cooked egg. In this regard, the molecular structure of the native, liquid protein was seen to be the greater barrier to digestion compared to the material properties of the cooked, denatured egg. It was also noticeable that the different material states of the cooked and raw egg contributed to significant differences in gastric emptying time. In the case of the raw egg, the gastric half emptying time was determined as 25 min (±9), whilst the cooked egg resulted in a pronounced increase in half empting time at 68 min (±6). Whilst it can be reasonably argued that the faster rate of emptying of the raw egg was due to its liquid state, it may also be the case that the poorer extent of digestion may result in reduced hormonal cholecystokinin (CCK) feedback signalling as a controlling factor in the rate of emptying when compared to the more readily digested cooked egg (Liddle, 1995).
A similar study, but under in vitro conditions was carried out by Luo and co-authors (Luo, Boom, & Janssen, 2015; Luo, Borst, Westphal, Boom, & Janssen, 2017) studying the peptic digestion of egg and whey protein solutions and comparing their behaviours to gelled systems comprising the same concentration of protein. Interestingly, they observed that the material properties of the gelled systems provided greater resistance to hydrolysis compared to the protein solutions. For the protein solutions, after 3 h gastric incubation, the extent of hydrolysis was determined as 11% and 15% for the egg and whey respectively. In contrast the extent of hydrolysis of the gels was found to be 2.5% and 7.9% for the egg and whey protein respectively. The reduced degree of hydrolysis for the gel systems was considered a consequence of limited diffusivity of the enzyme into the interior of the gel structure, and accordingly hydrolysis progressed primarily at the surface of the gel structures (Luo et al., 2017). The difference in observed effects when compared to the Evenepoel study (Evenepoel et al., 1998) may be due to the preparation of gel particles themselves, noting that the in vitro study was based on gel particles of cylindrical particles with dimensions approximately 5 × 5 mm. In contrast, the gelled egg in the human study would have undergone mastication during the eating process, and whilst the size of the particles was not reported, it might be reasonably assumed that these particles were both smaller and more structurally damaged as a consequence of the eating process, thereby enabling greater diffusivity of the proteases during digestion.
The role of food gel microstructure and assembly on subsequent digestive behaviours has been studied for other food proteins, indicating that the nature and strength of interactions can influence the manner in which the protein component is digested. This is exemplified in an in vitro study carried out by Rui et al. (2016) investigating the effect of different coagulants on the rate and extent of gastrointestinal proteolysis of tofu. Findings showed that under gastric conditions digestion was proceeded most rapidly for the acid coagulated gel, which displayed a higher overall extent of proteolysis. Digestion rate and extent was least for the tofu samples prepared through covalent cross-linking using transglutaminase. An interesting aspect of the research was the lack of correlation of digestion with material properties, noting that the acid tofus were significantly firmer than those prepared using divalent cations as a cross-linker. Indeed, it was observed that increasing the concentration of GDL used to coagulate the acid tofu was found not only to increase the firmness of the tofu, but also lead to a greater extent of proteolysis. One argument for the observed differences might be that acid tofu was at a lower initial pH, and therefore more favourable for pepsin activity compared to the other samples; however, the in vitro model used standardised the pH conditions at pH = 2 for all samples prior to addition of the pepsin. In the case of the samples comprising transglutaminase, the reduced level of proteolysis is consistent with observations made in Sect. 3.2 indicating that intramolecular and intermolecular cross-linking may limit the availability of hydrolysis sites for enzymatic binding.
Differences in protein digestion pathways can also be observed for other structured protein assemblies, such as demonstrated by Barbé and co-authors (Barbe et al., 2014). Their study investigated the digestion of two dairy gel systems of equivalent protein concentration but formed via different structural pathways, namely acid gelation and rennet gelation. Employing an in vivo pig model, they demonstrated that the acid milk gel was more readily proteolysed than the rennet gel. Additionally, significant differences in plasma peptides were observed, with the acid gel displaying elevated levels relative to the rennet get. These behaviours also manifested other variations in physiological biomarkers, for example in the release of ghrelin and CCK (although there was apparently no consequence of these variations on rate of outflow from the stomach). A number of reasons for these differences were postulated, noting that variations in sample pH (and the possible contribution to buffering effects in the stomach) might lead to differences in pepsin activity, with the acid gel enabling higher relative activities, at least during the early stages of digestion. Additionally, the two gels showed varying structural behaviour in the stomach, with the rennet gel undergoing extensive syneresis and consequential compaction of the protein network. This structural densification was considered inhibitory for diffusion of protease enzymes into the interior of the gel structure.
This particular hypothesis was explored further in considering how pore structure in protein particle networks might impact on pepsin enzymatic diffusivity (Thevenot, Cauty, Legland, Dupont, & Floury, 2017). Findings demonstrated that increasing the tortuosity and decreasing the pore size of protein gels through increasing protein concentrations did indeed reduce the ability of the enzyme to diffuse within the gel structure. It is important to note that such observations are also dependent on the overall structure dynamics of the gel system in response to the pH, temperature and shear conditions in the gut, and thus multiple dependencies are likely to determine the overall manner in which the protein is digestion.
In this regard, research to date is now starting to progress towards the complexities of protein digestion in multicomponent food systems. Findings are increasingly corroborating the hypothesis that the digestive breakdown pathway of any given protein component in a food will be determined by an integration of interactions across all length scales, from molecular to material, that are known to be influential in the breakdown of protein molecules during gastric and small intestinal residence. Arguably, and as observed with other macronutrients, human digestion and utilisation of dietary protein is remarkably efficient regardless of the type of protein or its structure. However, it is clear that there are variations in the digestive pathways of proteins based on their type, interaction within a formulation, structure and material properties and the response of these to the conditions in the GI-tract. Such variations may impart particular physiological outcomes, such as the management of satiety, or potentially have consequences for the manner in which other macronutrients and micronutrients are digested. Accordingly, there remains considerable research interest in exploring such relationships.
4 The Influence of Food Structure on Lipid Digestion
The primary source of dietary lipids comes in the form of triglycerides, which are naturally present in a variety of raw materials, of both plant (e.g. nuts, seeds and certain legumes) and animal origin, as well as being formulated into many processed foods. The apolar nature of triglycerides and their poor solubility in aqueous media presents a compositional state in food that is not immediately utilisable by the human body. Thus, the digestive process for lipids focusses on the conversion of dietary triglycerides into fatty acids and monoglycerides via enzymatic hydrolysis in the stomach and small intestine. These component products of lipolysis have a higher degree of polarity, and in combination with secreted bile salts and cholesterol, can be associated into nano-structured assemblies (the so-called mixed micelles), that are small enough to interact with, and diffuse across the epithelial membrane, a process that takes place in the small intestine. The nutritional value of lipids is not simply in the provision of fatty acids (acknowledging that whilst the body can synthesize most of the fats it needs from a range of dietary lipid sources, two essential fatty acids, linoleic and alpha-linolenic, cannot be synthesized in the body and must be directly obtained from foods in which they are naturally present) but also in acting as delivery vehicle for a range of lipophilic micronutrients. As indicated in the introduction, the role of fat is not just to act as a medium for ingestion of such micronutrients but also provides a critical role in their uptake and assimilation by the body during digestion (Agustiana et al., 2010; Jeanes et al., 2004).
4.1 The Role of Fat Structure in Food Systems
Lipid composition can vary widely according to source material; however, lipids tend to be generically classified in accordance with their physical properties at ambient temperature, that is, fats, such as butter, palm and coconut being solids under ambient conditions; and oils, such as sunflower, olive and canola, being liquids. Fats and oils play a significant role the hedonic aspects of foods, contributing to the temporal dynamics of lipophilic flavour perception (Arancibia, Jublot, Costell, & Bayarri, 2011; de Roos, 2006), arguably recognised as a tastant (Kindleysides et al., 2017; Running & Mattes, 2016; Stewart et al., 2010), and eliciting a broad array of textural descriptors, such as oiliness, greasiness and creaminess, amongst others (Scholten, 2017). This gamut of sensory characteristics is achieved through a remarkable diversity in the structural assembly of lipids within both natural and manufactured foods.
Within the context of lipid structuring in foods, we can broadly classify lipid structures as either non-colloidal, colloidal continuous or colloidal dispersed. Non-colloidal foods encompass a range of (generally low moisture) food materials for which the fat or oil is in a relatively free state, such as the presence of visceral fat on meat, or the use of oil for preparation of fried foods. Colloidal fat continuous foods are typically characterised by products such as butter and margarine (for which the dispersed phase is water), and in chocolate (for which the dispersed phase is solid particles of sugar and other non-fat cocoa solids). Colloidal fat or oil dispersed foods are probably the most predominant class of fatty food products, occurring naturally, as part of the cellular structure within many plant materials (such as oil bodies, within nuts, grains and seeds), or as emulsion droplets within mammalian milks. Colloidal fat dispersed structures are also widely encountered in manufactured liquid or soft solid food products, such as ice cream, mayonnaise and cheese. These emulsified systems are usually produced through the use of homogenisation to form typically micron-sized droplets, and through interfacial stabilisation of these droplets through adsorption of a surface active component within the formulation, either an ingredient (e.g. milk proteins in the manufacture of ice cream) or an additive, specifically added to the formulation for the purposes of emulsification (such as the use of gum Arabic in the stabilisation of flavour emulsions).
Arguably, the main purpose of emulsification is to provide a means for enabling a kinetically stable structure for dispersing lipids in an aqueous environment. This is exemplified in a number of foods, such as UHT milks, cream liqueurs, and liquid infant formulae, for which the formation of small stable droplets is able to maintain a homogeneously dispersed emulsion state for shelf lives in excess of a year. However, it should also be noted that controlled destabilisation of food emulsions has been used extensively in many products. This is particularly evident in soft solid (weak gel) and solid (strong gel) emulsions, where the material properties of the food are able to provide kinetic trapping of emulsion structures, thereby inhibiting phase separation over the product lifetime. Manipulation of droplet interactions into various states of aggregation, such as flocculated or partially coalesced is able to assist in the creation of soft solid or solid material states, and as well as assisting in the physical stabilisation of the system, can provide a significant contribution to the material, technical and sensory properties of the food.
Examples of this include the formation of protein-fat flocculated networks in the manufacture of yogurt and cream cheese, which provides a mechanism by which the firmness of the product can be enhanced, and the structuring of fat in ice cream through partial coalescence, which has been shown to markedly reduce the melting rate of the product. A similar partial coalescence mechanism that occurs in the manufacture of cheese, provides an important contribution to both the material properties of the cheese and associated characteristics, such as oiling off during melting. The ability to understand and control the structure of food emulsions has become an integral part of food design and development for many emulsified foods, with the ability to manipulate colloidal interactions and their spatial assembly within food structures enabling a degree of predictable control over product properties.
4.2 Motivations for Manipulating Fat Digestion
Naturally a question as arises as to whether our ability to digest and fully utilise dietary lipids is impacted by the structural diversity of fats and oils. In this regard, the human physiological pathway demonstrates a remarkable ability to process ingested fats and oils with equal efficacy regardless of their material or structural properties. This appears to apply irrespective of whether the lipids structures are present within natural food materials, or as part of manufactured products (noting that the consumption of processed emulsified foods forms a relatively recent component of our diet in terms of human evolution). Indeed, for healthy adults, the digestive efficiency of food lipids is typically in the order of ~95% for, when consumed as part of what can be considered normal dietary intake of fats and oils.
In recent years there has been significant intellectual investment directed to understanding how the lipid structures present in food impact on subsequent lipid digestion and digestibility (Mao & Miao, 2015). Such understanding has led to the wider consideration as to whether the processes of fat digestion can be manipulated to achieve a particular physiological outcome. Much of this research has been undertaken from a perspective of developing strategies for mitigating the current obesity crisis affecting significant populations across the planet. This includes the exploration of approaches capable of inhibiting or negating lipid uptake during digestion, and thus reduce the caloric intake associated derived from the lipid component of a food. Notable approaches to this effect include the utilisation of lipid-like materials which are able to impart equivalent technical and sensory properties relative to fats and oils, but which possess molecular structures impervious to digestion. Arguably, the most recognised of these materials are the sucrose polyesters (also known by the trade name Olestra) (Bimal & Zhang, 2006), which are essentially sucrose molecules with fatty acid chains linked through ester bonds at available hydroxyl sites. Liberation of these grafted fatty acids is not achievable through digestive lipolysis, and as the molecular structure of olestra is cannot be metabolised, it remains undigested across the entirety of the GI tract.
A slightly different approach involves the blocking of the digestive enzymes responsible for conversion of triglycerides to fatty acids, such as through the use of tetrahydrolipstatin (also known as orlistat) (Heck, Yanovski, & Calis, 2000). Orlistat acts by binding covalently to the serine residue of the active site of gastric and pancreatic lipases, disabling the ability of the enzymes to interact with the substrate (Guerciolini, 1997). As there is no mechanism by which non-hydrolysed triglycerides can be transported across the small intestinal epithelium, inhibition of hydrolysis means that varying amounts of ingested fats and oils can remain undigested. Both approaches provide mechanisms by which dietary fat intake can be lowered, and whilst at first glance it might be considered that such effects would be of benefit in developing strategies for mitigating the ongoing obesity pandemic, there remains an issue that the role of food is to provide nutrients for utilisation body. As such, for fats and oils for which digestive efficiency is particularly high, any mechanisms that lead to malabsorption to can cause physiological side effects (Harp, 1998). This includes concerns regarding inhibition in lipophilic micronutrient uptake (Borel, Caillaud, & Cano, 2015), potentially leading to deficiencies over extended periods, as well as issues relating to the excretion of non-digested (or non-digestible) lipids, leading to abdominal discomfort and steatorrhea (loose stool formation and anal leakage) (Kelly et al., 1998). Somewhat paradoxically, such strategies may actually promote greater food intake, with some studies indicating that any inhibition of fat digestion can result in increased appetite (Ellrichmann et al., 2008).
The complete inhibition of fatty acid uptake as detailed above represents something of an extreme approach towards manipulating lipid digestion. However, extensive research has shown that the dynamics of fat structure during digestion can be more subtly manipulated to achieve particular physiological responses. These include the ability to influence satiety signalling through particular pathways, such as controlling the rate of gastric emptying (Marciani et al., 2009), or by triggering the so-called ileal break mechanism (Madadlou, Rakhshi, & Abbaspourrad, 2016). Other potential outcomes include the ability to influence the oxidative stresses associated with fat digestion (Michalski, Vors, Lecomte, & Laugerette, 2017), enhancing the efficacy of lipophilic micronutrient bioavailability and lowering cholesterol levels (Chen, McClements, & Decker, 2013). In this regard, whilst such approaches generally conclude that it is actually quite difficult to manipulate the overall extent of lipid uptake associated with consumption of fats or oils, it is possible to exert some control over the rate and location of digestion. It should be noted that there is now extensive coverage in this field, and that collated findings are well represented in a number of excellent review articles (Golding & Wooster, 2010; Mao & Miao, 2015; McClements, Decker, & Park, 2009; Singh, Ye, & Horne, 2009; van Aken, 2010).
4.3 Principles of Gastric Fat Digestion, and Structural Pathways for Controlling This
The digestion pathway of lipids can again be influenced across varying length scales, from molecular through to material, and the dynamic response of these various length scales to the conditions encountered from point of ingestion and subsequent transit through the GI tract. However, the colloidal state is arguably of greatest consequence in ensuring effective lipid digestion. The poor miscibility of fats and oils in aqueous media requires the hydrolysis of triglycerides to be mediated through the adsorption of lipases at the oil–water interface. This facilitates the formation of nano-structured mixed micelles comprising a number of amphiphilic moieties including the hydrolysed fatty acids and monoglycerides, as well as bile salts, phospholipids and cholesterol. Interaction of these micelles and the brush border of small intestinal enterocytes enables diffusion of these structures across the membrane and into the epithelial cells (Lentle & Janssen, 2011).
Ignoring the debatable role of lingual lipase, lipid hydrolysis primarily occurs in the stomach via an acid stable gastric lipase, and in the small intestine via a bile salt dependent lipase–co-lipase complex. Gastric lipase is an acid stable enzyme that is secreted by the gastric chief cells in the fundic mucosa in the stomach. It has an optimum pH range of 3–6 and its amphiphilic structure allows for effective adsorption and positioning of the active site at the oil–water interface. Gastric lipase partially hydrolyses triglycerides to one fatty acid and one diglyceride, and is estimated as providing around 10–30% lipolysis in adults (whilst accounting for up to 50% total lipolysis in neonates, due to immaturity of the pancreas). Lipid digestion is continued in the small intestine via action of pancreatic lipase. Pancreatic lipase is the primary lipolytic enzyme during digestion, hydrolysing triglycerides at the 1,3 position to liberate two molecules of fatty acid and an Sn-2 monolgyceride. It is less effective at adsorbing at the oil–water interface than gastric lipase, and activity is greatest in the presence of co-lipase and where the adsorption of surface active bile salts to the oil–water interfacial layer has taken place (it can be argued that a key role of gastric lipolysis is to promote the occupancy of fatty acids at the oil–water interface, thereby facilitating the initial adsorption of the bile–lipase–co-lipase complex). The activity of pancreatic lipase can vary according to the presence of bile, but under physiological conditions is optimum at slightly alkaline pH. Furthermore, the adsorption of bile salts at the oil–water interface appears not only to enable pancreatic lipase adsorption, but is considered critical in the assembly of the mixed micelle moiety.
Lipolysis during both gastric and small intestinal transit represents an integral aspect of lipid digestion, and one that provides the most likely opportunity for manipulation. Indeed, much of the research focus in recent years has been the investigation of mechanisms by which the rate and extent of lipolysis can potentially be influenced. The previous section has indicated one such approach, in the use of drugs such as orlistat that specifically bind to the enzymes, effectively disabling their catalytic functionality to hydrolyse triglycerides. Whilst the side effects of enzyme blocking have been reported, it is noteworthy that this approach can be effective regardless of the structural or compositional state of the ingested lipid. In contrast, other approaches capable of influencing lipolysis are more reliant on manipulation of the colloidal state to achieve an effect.
The first of these is the creation of interfacial layers capable of acting as barriers to lipase adsorption. This concept tends to be most readily applicable to processed foods, such as milks, yogurts and ice cream, which incorporate oil-in-water emulsions, and for which the interfacial layer can be designed and manipulated through appropriate formulation and process control. The stabilisation of commercially produced food emulsions is generally achieved through two main classes of emulsifier: high molecular weight biopolymeric species or low molecular weight surfactants (a third option, that is, Pickering stabilisation via particulate adsorption has received considerable attention in the scientific literature but is not currently an established approach in product manufacture).
High molecular weight biopolymeric materials typically encompass a broad range of proteins, most commonly milk proteins (and for which there are multiple ingredient options provided by suppliers), but also increasingly from non-dairy sources such as soy, wheat, egg, pea and rice, amongst others. In addition, a number of additive systems capable of stabilising emulsions (generally derived from polysaccharides) are commercially available. These include gum Arabic, acetylated pectins and a number of chemically modified materials, such as OSA starch, propylene glycol alginates and hydrophobised cellulose derivatives (e.g. hydroxypropyl cellulose).
Low molecular weight surfactant species are based specifically on edible lipids, and tend to have well defined amphiphilic characteristics in which the hydrophobic domain is based on a fatty acid chain (which can vary in length and degree of saturation) and for which the acid group is esterified with a polar or charged moiety which provides the hydrophilic aspect of the molecule. These materials are generally classed as additives, and are more commonly termed as emulsifiers, which can be misleading as their technical function in food extends beyond just the stabilisation of droplets. Phospholipids represent the most naturally derived class of low molecular weight surfactants, and can be structurally summarised as two fatty acid chains linked to a phosphate head group. Collectively termed as lecithins, phospholipids can be obtained from a number of dairy and plant sources, including milk, eggs and seeds (e.g. sunflower and canola). Depending on source material, phospholipids can show broad structural variation in both fatty acid composition and in regard to the side-groups attached to the phosphate head group, which in turn influences their functionality as used in food products. Commercially produced lecithins typically comprise a mix of phospholipids reflective of the raw material used in their production. Chemically synthesised polar lipids are also widely utilised within the food industry, in which fatty acids (usually based on palmitic, stearic or oleic fatty acids chains) are chemically esterified with a head group containing at least one reactive hydroxyl group. Monoglycerides, which are produced through reaction of triglycerides or fatty acids with glycerol, are the largest class of chemically manufactured class of emulsifiers. Further variations in head-group type are able to produce an extended range of edible surfactant systems, which are broadly classified according to the relative contribution of hydrophobic and hydrophilic domains within the molecular structure (the so-called hydrophilic–lipophilic balance, or HLB). The specific properties of emulsifiers can vary considerably according to the choice of fatty acid and head group, and thus this particular class of food additives is able to provide very broad, yet specific, technical function in food production, including the stabilisation of oil-in-water emulsions.
Both biopolymeric and low molecular weight surfactants are extensively used in the stabilisation and structuring of food emulsions, noting that the particular choice of emulsifier, its behaviour during processing, and its interactions with other components within a formulation provide a determining effect on the stability, structure and properties of the emulsion. As such, there is considerable interfacial and structural variety across the range of emulsified foods that we consume. The particular interfacial composition of food emulsions can influence the pathway by which these materials are processed during gastrointestinal transit; however, as indicated earlier it appears that our physiology is well equipped to manage different interfacial states and structure, given the almost complete ability to uptake dietary lipids when integrated across the entirety of the stomach and small intestine.
The digestion of food emulsions and the particular role of interfacial composition have been widely explored, from model systems to actual foods and from in vitro characterisation through to human trials. In considering the gastric stage of digestion, notable differences are observed depending on whether emulsified lipids are stabilised with proteins, polysaccharides or surfactants. If we first consider the stabilisation of emulsion systems with protein (which is arguably the most widely utilised approach for formulating soft solid and solid emulsion-based foods), it becomes apparent that such systems can be particularly sensitive to the pH dynamics in the under gastric conditions. In the fasting state, the pH of the stomach is typically 1.9–2.0. However, this can rise significantly on ingestion of food due to the buffering effects of the food system, and this is particularly evident for protein-based compositions. In the case of protein-stabilised emulsions, the initial elevation of gastric pH is followed by gradual lowering back towards fasting levels, which can effect a transition through the isoelectric point for some protein systems, such as the caseins (van Aken, Bomhof, Zoet, Verbeek, & Oosterveld, 2011) and whey proteins, and which can lead to rapid flocculation of emulsion droplets in the early stages of gastric incubation (Bellesi, Martinez, Ruiz-Henestrosa, & Pilosof, 2016). The stability of protein stabilised emulsions can be further compromised during residence in the stomach as a consequence of proteolysis of the interfacial layer by pepsin.
The consequences of peptic hydrolysis on protein coated interfacial layers and emulsion properties have been explored in a number of studies. At the planar oil–water interface, Maldonado-Valderrama and co-authors (Maldonado-Valderrama, Gunning, Wilde, & Morris, 2010) showed how proteolysis under gastric conditions affected the structural and mechanical properties of a β-lactoglobulin adsorbed layer, demonstrating that whilst partial hydrolysis of the interfacial protein layer took place leading to a reduction in surface tension, complementary atomic force microscopy showed that, to a degree, the interconnected interfacial network remained intact. Accordingly, the dilational elasticity of the interface was only partly lowered as a consequence of enzymatic digestion. The retention of the interfacial network was speculated to be due to strong hydrophobic interactions, not only between protein and the interface but between neighbouring peptide fragments.
In the case of a model β-lactoglobulin-stabilised emulsion, Sarkar et al. (2010a) showed that in vitro exposure of the emulsion systems to gastric fluid at pH 1.2 and containing pepsin resulted in a time-dependent reduction in zeta potential from +50 to +17.6 mV during a 2-h incubation. This reduction in zeta potential was attributed to the detachment of charged domains of the protein layer from the interface. SDS-PAGE measurements were additionally used to show that after 2 h of digestion only ~20% of the interfacial protein membrane remained intact.
Proteolytic removal of the electrostatic and steric stabilising layer can also lead to flocculation of protein-stabilised emulsions (Hur et al., 2009; Golding et al., 2011), promoting droplet association through increased hydrophobic interactions (as observed, this may occur in the later stages of gastric incubation, compared to flocculation as a consequence of pH effects). A further consequence of the interfacial proteolysis is a reduction in the mechanical strength of the interface (Maldonado-Valderrama et al., 2009). Thus, for flocculated, weakly stabilised emulsion droplets, even mild shear conditions can be sufficient to induce coalescence. For liquid-oil-stabilised emulsions this has been shown using microscopy in a number of separate studies (Hur et al., 2009), which has shown coalescence susceptibility as a consequence of prolonged exposure to the simulated gastric environment. Both flocculation and coalescence of may lead to a phase separation of the emulsion during gastric digestion.
It should also be noted that the initial interfacial proteolysis of protein stabilised emulsions may facilitate the adsorption of gastric lipase at the oil–water interface by creating hydrophobic domains that provide preferential adsorption sites for the lipase enzyme to bind to (Lueamsaisuk, Lentle, MacGibbon, Matia-Merino, & Golding, 2014, 2015). As indicated earlier, gastric lipolysis only accounts for approximately 10–30% of total synthesis of fatty acids. However, this initial lipolytic step can itself result in dynamic changes to the composition of the interfacial layer during gastric incubation, due to the accumulation of surface active fatty acids at the oil–water interface (Pafumi et al., 2002b). The build-up of fatty acids at the interface ultimately becomes inhibitory to lipase accessibility to the interface (such that lipase action effectively ceases at surface fatty acid concentrations of between 110 and 120 μmol m−2).
Whilst there are a number of non-protein biopolymeric emulsifiers capable of stabilising food emulsions, these are less widely encountered in actual manufactured food products. The use of gum Arabic as an emulsifier for the production of encapsulated bioactive materials or in the stabilisation of lipophilic flavour emulsions provides specific examples, and indicates some of the benefits in using such materials compared to proteins, such as maintenance of physical stability over a wide range of pH, ionic and thermal conditions. For these same reasons, non-protein biopolymer stabilised emulsions can display distinctly different behaviours during gastric incubation when compared to protein stabilised emulsions. Notably, adsorption of particular polysaccharide layers can provide effective steric stabilisation under the acidic conditions encountered in the stomach, and therefore do not undergo pH mediated flocculation. This is exemplified by research undertaken by Bellesi and co-authors (Bellesi et al., 2016), who demonstrated that emulsions stabilised with hydroxpropylmethyl cellulose, a non-ionic cellulose derivative, did not undergo significant destabilisation during gastric in vitro treatment when compared to similar emulsions stabilised with either soy protein or whey protein isolate. In addition to resistance to flocculation under gastric pH, it was also observed that polysaccharide stabilised interfaces were impervious to peptic hydrolysis, which provide an additional mechanism for maintaining physical stability. Accordingly, consumption of such emulsions may lead to a more homogeneous distribution of droplets within the stomach during gastric residence which, as will be discussed, can have implications on gastric emptying rate. However, it is less clear, at least under in vitro conditions, as to whether such interfacial layers are in any way inhibitory to gastric lipase adsorption. This is primarily due to the current limited availability of human gastric lipase (or appropriate mammalian alternative) for incorporation into in vitro models. The use of various fungal lipases has been employed in some models, but these can lack physiological equivalence, with variations in active site location, specificity and pKa.
Whilst the above examples give consideration to the use of various amphiphilic gums as emulsifiers, it should be noted that the majority of food grade polysaccharides are hydrophilic and thus not surface active, and cannot therefore be used directly for the stabilisation of emulsions. However, it is possible to attach polysaccharides to protein coated droplet through various binding mechanisms, most commonly electrovalent or covalent cross-linking or combinations of the two. Interfacial layers comprising protein-polysaccharide conjugates have also been shown to impart effective droplet stabilisation when exposed to pH conditions representative of the stomach (McClements & Li, 2010; Tokle, Lesmes, Decker, & McClements, 2012).
4.4 Fat Partitioning in the Stomach and Gastric Emptying
Variations in interfacial composition show that the physical stability and extent of lipolysis of food emulsions within the stomach can differ considerably according to formulation and thus be manipulated. Whilst such variations tend not to have significant consequence in altering the total lipid uptake from a food or meal, there can be some interesting differences relating to the kinetics of digestion, notably the specific influence of fat partitioning within the stomach on gastric motility and the rate at which lipids are released into the small intestine. The relationship between fat distribution in the stomach and gastric emptying has been established using model beverage emulsions in a number of separate in vivo studies. This includes research by Marciani and co-workers (Marciani et al., 2009; Marciani, Wickham, Bush, et al., 2006; Marciani, Wickham, Singh, et al., 2006; Marciani et al., 2007), who demonstrated how the partitioning of the fat phase in the stomach affected emptying. Consumption of 500 ml beverage emulsion systems that were designed to either separate or remain stable in the stomach were used to explore how fat separation influenced rate of emptying. The distribution of fat in the stomach was visualised using magnetic resonance imaging (MRI), which was able to clearly distinguish between the gastric-stable and gastric-separating compositions. For these model emulsion formulations, gastric emptying was observed to be initially more rapid for consumption of gastric-unstable emulsions, due to release of the oil-depleted aqueous phase into the intestine, with the emptying rate significantly decreasing on release of the oil-separated portion into the intestine. In comparison, the gastric-stable emulsion displayed a slower and more uniform rate of emptying, due to consistent release of fat from the stomach during the emptying period. Differences in emptying behaviour showed correlations with CCK response, such that the gastric unstable emulsion induced significantly lower measured plasma CCK in comparison to the gastric-stable emulsion, for up to 6 h after initial consumption of the emulsion. Data from satiety assessment also indicated that the acid-stable emulsion was more effective at suppressing hunger.
Further research comparing separated and stable emulsion systems across a range of fat and macronutrient concentrations showed this to be a reproducible and controllable phenomenon (Foltz et al., 2009; Keogh et al., 2011).The specific emptying behaviour of the layered emulsion is attributed to a two-stage mechanism with rapid emptying of the oil-depleted aqueous portion of the meal. With little or no delivery and detection of fat in the intestine, there is minimal secretion of CCK, and accordingly emptying progresses rapidly. Once the fat layer leaves the stomach and enters the intestine there is a corresponding increase in CCK which slows down the rate of emptying. In the case of the emulsified beverage, commencement of gastric emptying results in immediate delivery of lipid into the intestine. The corresponding stimulation of CCK then regulates the rate at which meal contents is released from the stomach and this feedback mechanism proceeds according to the continuing detection of fat on emptying. In comparison to the layered emulsion, this results in a more uniform emptying rate, with markedly slower emptying in the early stages of digestion.
Arguably, much of the research conducted on lipid digestion, including the examples provided thus far, applies primarily to stable, non-interacting emulsions prior to exposure to gastric conditions. Thus, in terms of food formats, the model systems presented within these studies would mostly be considered representative of beverage formulations, such as milks, infant formulae, liquid dietary supplements and even ice cream (which can be considered liquid post-consumption).
In contrast to many of the systems discussed above, non-colloidal fatty foods and for fat continuous emulsion systems do not typically enter the stomach in the form of finely emulsified droplets. The markedly low surface area of such foods might be expected to reduce the rate and extent of gastric lipolysis, as this will greatly reduce the availability of binding sites for lipase adsorption compared to finely dispersed oil–water emulsion systems for which the surface area may be orders of magnitude higher. The surface area of unstructured fatty foods can be increased through oral processing, noting that the mechanical action in mouth coupled with mixing with saliva can enable the crude emulsification of fats into large droplets (Adams, Singleton, Juskaitis, & Wilson, 2007). Likewise, the eating process of fat continuous foods such as chocolate or butter can lead to the phase inversion of the initial structure during oral residence, resulting in the formation of oil-in-water type structures, albeit with large droplet size distribution (Rodrigues et al., 2017). Even crude homogenisation can greatly increase surface area compared to a fully phase separated oil layer, which can assist with the onset of lipolysis. It should be noted however that the large size of such droplet will still most likely result in some partitioning of the lipid phase towards the stomach, which as indicated may lead to differences in rate of emptying compared to finely dispersed gastric stable compositions.
For semi-solid or solid protein stabilised emulsion foods (such as cream cheese, Greek yogurt or processed cheese), the composition and structural state of the food causes greater complexity in the way in which these systems behave during digestion. A complicating factor is that whilst the structure of liquid emulsions tends to be minimally impacted during consumption, the oral processing of soft solid or solid food can lead to a considerable alteration in structure. Most examples of such foods are in the form of oil-in-water emulsions; however, there can be considerable variations in droplet size, surface composition and interactions with the surrounding continuous phase. For example, many soft solid emulsions comprise an appreciable amount of protein, which can contribute to the structure as well through formation of the type of network structures discussed in Sect. 3.4. Given that the kinetics of lipolysis are determined by the ability of the lipase to access the droplet surface, such structures may initially be inhibitory to lipase adsorption. The mechanical action in the stomach and the concerted action of pepsin will invariably lead to a deconstruction of the ingested food structure, but the onset and rate of lipolysis may be slower than that of non-structured emulsions (Wooster et al., 2014).
4.5 Lipid Digestion in the Small Intestine
Lipolysis in the small intestine continues to completion through adsorption of co-lipase-dependent pancreatic lipase, which hydrolyses triglycerides at the sn-1 and sn-3 positions to liberate two fatty acids and an sn-2 monoglyceride. Pancreatic lipase is active under the neutral to mildly alkaline conditions found in the small intestine, with optimal activity at ~6.5. Whilst able to catalyse lipolysis through direct adsorption to an interface, its adsorption is more readily inhibited by the presence of any surface-active components already occupying the interface. Accordingly, lipolysis efficiency is greatly enhanced by the secretion of co-lipase and bile salts (Blackberg et al., 1979). Co-lipase is a non-enzymatic protein co-factor which is more amphiphilic than pancreatic lipase, and is able to form complexes with the C-terminal region of the enzyme. This complexation provides a more favourable environment for adsorption at the oil–water interface (Lowe, 1997). However, lipase–co-lipase adsorption may still be inhibited by highly surface-active molecules adsorbed to emulsion droplets on entry to the small intestine. For example, stabilisation of an emulsion system with the small molecule surfactant Tween 80 was shown to render the interface impervious to both pancreatic lipase and complexed lipase–co-lipase (Gargouri, Julien, Bois, Verger, & Sarda, 1983).
Bile salts and phospholipid secretion by the liver can facilitate intestinal lipolysis of fats and oils (Maldonado-Valderrama et al., 2011). Whilst bile salts are amphiphilic, their structure is atypical of the usual head group–tail group molecular make-up of small-molecule surfactants. The basic structure of bile salts can be described as a rigid steroid backbone comprising hydrophobic and hydrophilic faces which is attached to a flexible region. Bile salts can be expected to possess greater surface activity than most edible amphiphiles used for interfacial stabilisation of fats and oils and will tend to displace these through a mechanism of orogenic displacement (Bellesi, Ruiz-Henestrosa, & Pilosof, 2014; Maldonado-Valderrama et al., 2008).
However, for polar lipid surfactants with greater relative surface activity than bile salts orogenic displacement is less likely to occur. Instead, it has been shown that for phospholipid monolayers, a synergistic interaction between the bile salts and the polar lipid interface allows for bile salts to be co-adsorbed at the surface, as evidenced by a lowering of interfacial tension beyond that of the surfactant system alone (Gallier, Shaw, et al., 2014).
The adsorption of bile salts at the interface is an important step in the digestive process, as it provides binding sites for the co-lipase–lipase complex at the interface, allowing lipolysis to proceed. This has been exemplified through investigation of the digestion of model emulsions stabilised by the non-ionic surfactant Tween 80, where it was observed that in the absence of bile salts, the lipolysis of the Tween stabilised emulsions was inhibited, whereas the inclusion of bile salts enabled the adsorption of the lipase–co-lipase complex, allowing lipolysis to proceed (Gargouri et al., 1983).
With this in mind, any interfacial mechanism that is able to inhibit adsorption of bile salts to the oil–water interface is likely to limit lipolysis. This has been demonstrated for oil droplets stabilised with non-ionic galactolipids. In comparison to charged phospholipids, galactolipids had reduced interaction with bile salts whilst imparting a more densely packed interface, thus restricting bile salt accessibility to the oil surface, and with a corresponding reduction in susceptibility to lipolysis (Chu et al., 2009). The formation of multilayered interfacial layers has also been shown to be effective against bile-salt adsorption, as exemplified by the formation of charge complexed bilayers comprising anionic lecithin and cationic chitosan, which was shown to have a reduced rate of lipolysis under in vitro intestinal conditions when compared to a control emulsion solely stabilised by lecithin (Mun, Decker, Park, Weiss, & McClements, 2006). However, whilst the effects of interfacial complexation clearly resulted in a modified lipolysis rate under in vitro conditions, replication of the study design in vivo (using a mouse study), did not appear to have any influence on key digestive biomarkers relating to fat uptake when compared to an unmodified control (Park et al., 2007).
In this regard, it should be noted that whilst (under in vitro conditions and potentially via direct small intestinal intubation) variations in interfacial composition as presented by these example can influence small intestinal lipolysis, it is less likely that for the majority of consumed foods, the interfacial composition is likely to be inhibitory to bile salt adsorption, and indeed it may be the case that the original interfacial composition of any ingested food colloid may have already been displaced by fatty acids as part of gastric lipolysis.
However, lipase accessibility to an interface is not the only mechanism by which emulsion lipolysis in the small intestine can be influenced. As stated, lipolysis is an interfacially mediated process, and thus the efficiency of enzymatic hydrolysis should be determined by the available surface area of the substrate, that is, the oil–water interface. On this premise, decreasing the size of emulsion droplets entering the small intestine should, in principle, lead to increased rate and extent of lipolysis due to the increased availability of lipase binding with increasing surface area. In vitro studies on model emulsion systems designed to undergo gastric destabilisation destabilisation do appear to show that the rate and extent of pancreatic lipolysis is suppressed as surface area is reduced (Golding et al., 2011), with emulsions displaying coalescence exhibiting lower levels of released fatty acids in comparison to stable emulsions retaining high surface area. Additional in vitro studies, in which the droplet size of the emulsions is controlled prior to exposure to simulated intestinal fluid, also show that lipolysis is correlated to surface area (Li, Hu, & McClements, 2011).
However, these findings do not necessarily correlate when applied in vivo. Certainly, where gastric digestion is bypassed and emulsions are delivered directly into the small intestine through intubation, a relationship between droplet size and lipid digestion efficiency does appear to exist. This was demonstrated by Seimon and co-authors (Seimon et al., 2009) who investigated the small intestinal digestion of model emulsions with droplet size ranging from 0.26 to 170 μm. Their results indicated that the highest surface area emulsions generated statistically significant elevated levels of CCK and PYY, as well as higher levels of plasma triglycerides. The finer emulsions also had a marked effect on intestinal motility, reducing intestinal transit rates to ensure full digestion and uptake of fat. A further study, also looking at the effects of direct intubation showed that increasing surface area influenced not only digestive biomarkers, but also relative food intake.
In determining a correlation between surface area and intestinal digestibility of fat, these studies intentionally ignored the role of the oral and gastric environments on the emulsion structure and digestion dynamics prior to entry in the small intestine. A notable in vivo study by Armand and co-authors (Armand et al., 1999) took a more integrated approach, employing intubation of two model emulsions of differing size (10 and 0.7 μm) into the stomach and monitoring changes to size distribution, gastric and pancreatic lipase activities and fat digestion. Lipolysis under gastric conditions was seen to be greater for the fine emulsion, consistent with the argument that higher surface area facilitates lipolysis. However, lipolysis was accompanied by an increase in droplet size during gastric incubation. A difference in lipolysis between the two emulsions was also observed in the duodenum; again, somewhat greater for the fine emulsions. However, the overall plasma triglyceride counts were not significantly different between the two emulsions, indicating that the overall lipid uptake was unaffected by the initial droplet size. One other point of consideration from this study was the fact that the peak point for plasma triglyceride was significantly delayed in the case of the fine emulsion, indicating that differences in fat distribution during gastric digestion may have affected the rate of stomach emptying of the two emulsions. This apparent normalisation of emulsion structure during digestion is perhaps understandable, given the biological requirement to achieve effective lipid nutrient uptake. A number of studies suggest that bile salt/phospholipid adsorption appears to be an important factor in regulating the colloidal state during the intestinal stage of digestion (Golding et al., 2011; Nik, Wright, & Corredig, 2011; Sarkar, Horne, & Singh, 2010b), all indicating that for fine emulsions (droplet size typically <1 μm), significant coalescence can take place during incubation in intestinal fluid containing bile salts (irrespective of surface composition).
4.6 Fat Digestion: Mechanical Factors Altering Digestion
Thus far it appears that human physiology is well equipped to ensure effective lipid digestion across most natural or processed foods that comprise the human diet. The combined pH, enzymatic and physical conditions across the mouth, stomach and small intestine are able to accommodate a broad array of colloidal structure with remarkable variations in size, surface area, interfacial and lipid composition, enabling full hydrolysis of triglycerides from these diverse lipid sources, and complete utilisation of the rendered fatty acids. The above sections indicate that lipid and emulsion based interfaces and structures can be designed that extend beyond normal digestive mechanics, leading to variations in digestive outcome, these are not widely encountered as part of food design. However, it should be noted that lipid digestion can be influenced by the mechanical properties of the food, and where any ingested food system is consumed such that its material properties are resistant to digestion, this can lead to variations in ability of the body to full process the nutrients present within those structures. This has been observed for both natural and constructed food materials.
Findings from this study and elsewhere (Grundy et al., 2015) imply that the oral processing of almonds is insufficient to fully disrupt the cellular structure, thus liberating the encapsulated oil bodies for digestion. Indeed, it is believed to be primarily the outer layer of the seed that is sufficiently masticated to enable access for lipase enzymes during gastric and small intestinal digestion, and that where disruption has taken place oil body coalescence can occur resulting in formation of liberated droplets in the order of 10–40 μm that are readily accessible for enzymatic adsorption (Ellis et al., 2004). Where cellular disruption has not occurred, the intact cell wall appears remarkably resistant to decomposition in either the stomach or small intestine, and thus oil bodies retained in these structures are not able to be hydrolysed and remain unavailable for transport across the epithelium. It has been postulated that some lipid leeching can occur during gastrointestinal transit, and that swelling of cellular structures may ultimately allow diffusion of digestive enzymes and bile salts into the interior of the cell at sufficiently long digestion times. It has also been observed that pectic fermentation by colonic microflora can eventually start to breakdown cell structures along with fermentation of the lipid component. However, it is clear that a considerable portion of the lipid component is not available as metabolisable energy and thus the utilisable energy content of almond seeds is lower than the total energy content (Novotny, Gebauer, & Baer, 2012). A further consideration is that the inhibited uptake of the oil content present in almonds extends to a reduction in uptake of lipophilic micronutrients (such as tocopherols or phenolics) that are associated with the lipid phase.
Any modification of the native structure through processing might be reasonably expected to impact on the digestibility of the material; accordingly, the impact of thermal processing (roasting) and mechanical disruption have been investigated in relation to the bioaccessibility of the lipid fraction of the almonds. A number of studies show that roasting appears to actually have limited impact on the availability of the lipid component when the almonds are consumed whole kernels (Bornhorst et al., 2013; Mandalari et al., 2014). Likewise, limited comminution of the intact seeds into a ground state has only a marginal effect of availability. In contrast, processing approaches that fully disrupt the cellular structure, thereby liberating the entrapped oil bodies, such as in the production of almond milk, have been shown to greatly enhance the viability of the droplets towards lipolysis, and thus greater uptake is observed (Gallier, Rutherfurd, Moughan, & Singh, 2014).
Increasing the mechanical structural resistance of lipids to digestion has also been observed in manufactured emulsion systems. The particular approach demonstrated by Golding and co-authors (Golding et al., 2011) explored the hypothesis that lipid digestion rate and extent could be decreased in accordance with a decrease in surface area of fat during the digestion process. To this end, the study design focussed on synthesising a number of model emulsion systems that could be dynamically structured under gastric condition, from stable to flocculated to coalesced, allowing control over surface area of the emulsion during digestion. Whilst in vitro measurements showed that rate of lipolysis and liberation of free fatty acids could be affected by changes in surface area, a corresponding human trial showed that plasma triglyceride concentrations were not significantly affected as a consequence of any structural changes occurring during digestion (although it was observed that the emulsion predicted to undergo gastric coalescence did demonstrate a delay in onset of plasma triglycerides, consistent with previous studies in which lipid separation in the stomach delayed the entry of lipids into the small intestine) (Golding et al., 2011; Keogh et al., 2011).
These findings applied when the emulsions comprised oils that were fully molten at in body temperatures. When reformulated using a fat blend that had a solid fat content of 25% at 37 °C, the digestion properties were more noticeably affected. This was particularly true of an emulsion system designed to undergo partial coalescence (i.e. droplet agglomeration) in the stomach, leading to the formation of visible lumps of aggregated fat. Under in vitro conditions the partially coalesced emulsion was shown to have a markedly reduced rate and extent of fatty acid liberation arising from lipolysis. When extended to the human study, a similar pronounced suppression of plasma triglyceride was observed. The interpretation of these findings suggested that whilst the surface area reduction of liquid oil droplets was ultimately limited by gastrointestinal emulsification, ensuring that a colloidal state could be maintained throughout the digestion process, the biomechanics of the GI-tract were insufficient to break up the partially coalesced aggregates of fat, and thus the low surface area was constant during digestion.
Interestingly, an additional study not only provided further evidence of this effect, but also determined that the interactions leading to partial coalescence during gastric digestion could be influenced by the continuous phase composition of the emulsion, leading to variation in rate and extent of fat digestion dependent on whether partial coalescence was able to take place in the stomach (Wooster et al., 2014). Whilst the findings demonstrated that this approach could modulate the rate of lipid digestion, the translation of this effect into food systems is still limited, since most dietary fats or oils (whether consumed as part of natural or processed food materials) tend to be fully molten at in body temperatures. A particular challenge of utilising this effect in a food product (acknowledging that a slower rate of lipid digestion may have beneficial effects in reducing lipaemic inflammation) is the potential alteration of textural properties associated when utilising a fat source that remains solid at in-body temperatures.
5 Effect of Structure and Composition on the Digestion of Carbohydrates
5.1 Digestion of Glycaemic Carbohydrates
5.1.1 Monosaccharides and Disaccharides
As with protein and lipids, the purpose of carbohydrate digestion is to render the carbohydrate component of food materials into a molecular state utilisable by the human body: in the case of carbohydrates, primarily as a source of metabolisable energy. Similarly, this process is predicated through biochemical translation of carbohydrate molecules through a suite of different carbohydrase enzymes, producing component monosaccharides capable of diffusing across the small intestinal epithelium. It is worth noting that certain dietary monosaccharides, namely glucose, galactose and fructose, are in a form already able to be transported across the epithelium, noting that whilst glucose immediately utilisable for metabolism by all tissues, the majority of fructose and galactose is metabolised in the liver (Englyst, Liu, & Englyst, 2007). Monosaccharide carbohydrates are highly water soluble, and in aqueous media (such as beverage formats, particularly those designed for sports nutrition) can be rapidly absorbed due to fast transit of liquids through stomach and small intestine, along with the lack of need for hydrolysis.
For foods with a higher level of structural complexity, the bioavailability and rate of uptake of the becomes more dependent on the liberation of the carbohydrate from the food matrix during digestion. This can apply to both natural food materials. For wholefoods, such as fruit and vegetables, the carbohydrate is solubilised within the cell structure of the material. For manufactured foods, the structural state of the carbohydrate can vary considerably, being present in liquid form, in products such as ice cream and yogurt, through to incorporation as a solid in low moisture compositions, such as confectionary, biscuits and cereal foods. Liberation of monosaccharides from these carbohydrate containing foods commences in the mouth, where oral processing is able to breakdown structures thereby exposing the carbohydrate to the oral cavity. Additionally, mixing with saliva is able to commence the solubilisation of solid state monosaccharides leading to dissolution of structures. These processes are invariably linked to temporal hedonic perception of sweetness (Arancibia, Costell, & Bayarri, 2013; Kohyama, Hayakawa, Kazami, & Nishinari, 2016).
It should be noted that the extent of release of soluble material will be partly dependent on the bolus forming properties of the consumed food. For example, food structures formulated predominantly with soluble carbohydrate components, such as confectionary products can be mostly disintegrated during the eating process, leading to extensive liberation of any monosaccharide component on entry to the stomach. Conversely, solid foods containing other macronutrient components may be broken down in the mouth, releasing a certain amount of carbohydrate prior to bolus formation. Subsequent bolus formation may, in turn, cause some reassembly of the components in the food, thereby reducing the extent of oral solubilisation. Likewise, for carbohydrates contained in plant tissue in fruits and vegetable, cellular breakdown during oral processing may release some of the encapsulated sugars; however, depending on the material properties of the food in question (which in turn may be governed by factors such as ripeness), the mechanics of eating may not complete breakdown all the cell structures leading to limited retention on entry into the stomach (Harker, Amos, Echeverria, & Gunson, 2006). On entry to the stomach, exposure to gastric fluids and mixing causes further release of carbohydrates from structured food systems—a process which can be aided by the hydrolysis of protein structures. For multicomposite natural or manufactured food material, the kinetics of food structure breakdown in the stomach can influence the rate of small intestinal uptake of monosaccharides (Southgate, 1995). This can be due to either the rate and extent of liberation of the carbohydrate component from the digesta during gastric incubation, or alternatively due to a variation in rate of gastric emptying arising from the breakdown and digestion of food structures and specific effects such as CCK regulation of emptying rate as a consequence of fat digestion (Rayner, Samsom, Jones, & Horowitz, 2001).
As indicated, for the monosaccharide carbohydrates glucose, fructose and galactose, digestive hydrolysis is not required to render these molecules in a transportable state across the epithelium. For other simple, soluble carbohydrate components, such as the range of disaccharides that can comprise our diet (most notably sucrose), a similar consideration in terms of release, and depending on structure, solubilisation of these materials during the eating and digestion process can influence their rate of uptake. However, these sugars require additional digestive hydrolysis to convert them into component monosaccharides that are compatible with epithelia diffusion. This hydrolysis step is located in the domain region of the small intestinal enterocytes that are correspondingly located proximally to the transporters which will carry the hydrolysed sugars into the epithelial cells. The enzymes responsible for this part of the digestion process (isomaltase, sucrase and lactase inter alia) are not unbound in the intestinal lumen, but anchored to the membrane proteins in the plasma membrane of the enterocyte. The apical plasma membrane housing these so-called brush border enzymes comprises microvilli, which protrude from the cell and constitute the brush border region (Hooton, Lentle, Monro, Wickham, & Simpson, 2015). Arguably the combined action of oral, gastric and small intestinal digestion ensures that most disaccharide carbohydrate is in a bioavailable state (i.e. fully solubilised and not trapped within digesta material) for hydrolysis by the brush border enzymes.
The third glycaemic carbohydrate fraction that is assimilated in the small intestine is starch, which forms a significant compositional and structural component of many natural and manufactured foods. Starch is a polymeric form of glucose, and is produced by many plant materials as an energy store, with staple crops such as corn, potato and wheat being notably starch-rich. The molecular composition of starch comprises two main types, the linear, unbranched amylose fraction and the branched amylopectin fraction. Generally, amylopectin is the more abundant of the two, typically comprising 70–80% by weight of most plant materials; amylopectins are also of higher molecular weight relative to the amylose fraction. The large size of starch molecules prevents direct transport across the epithelium and thus enzymatic hydrolysis yielding free glucose is the mechanism by which both amylase and amylopectin can be digested. This is achieved at two specific locations during digestion.
Starch digestion is initiated in the mouth by secretion of an α-amylase (which is also termed ptyalin) present in saliva (Butterworth, Warren, & Ellis, 2011). This enzyme has optimum activity under in-body conditions, that is, at a pH of 7, and temperature of 37 °C. Whilst salivary amylose can hydrolyse starch into the disaccharide maltose, it is not common for this to happen during eating (as evidenced by the fact that starchy foods do not generally tend to increase in sweetness during oral residence). Oral processing can be sufficient for initial conversion of starch into oligosaccharide fraction as well as enhancing the solubilisation of insoluble starch materials. Whilst salivary amylase is inactivated in the acidic conditions in the stomach, in reality, inactivation can be delayed as a consequence of any pH buffering effects generated by an ingested food capable of retaining gastric pH levels elevated above that required for amylase inactivation (noting that this can be particularly evident in foods comprising high levels of protein). Additionally, bolus formation may lead to entrapment of amylase within the bolus structure. Slow diffusion of gastric juices into the bolus may reduce the rate of inactivation of the entrapped enzyme (Mennah-Govela, Bornhorst, & Singh, 2015), allowing hydrolysis to be continued within the stomach.
Starch hydrolysis is continued in the small intestine through action of pancreatic amylase which randomly cleaves the glycosidic bond to progressively reduce starch to oligosaccharide fragments and ultimately disaccharide units of maltose. The maltose is then further acted upon by the brush border enzymes maltase and isomaltase, yielding free glucose that can in turn be transported across the epithelium.
This has been observed with digestion of pastas, in which the gelatinised starch-protein network forms large fragments during oral processing that are relatively inhibitory to amylase diffusion during gastrointestinal transit (Thomsen et al., 1994). This is also noted to occur in cooked beans (e.g. baked beans), in which the starch component is present within the cellular structure of the bean. Cellular structures damaged during mastication enable accessibility for the amylases, however, where cell structures are undamaged, the encapsulated starch can be retained in an unswollen state due to spatial limitations, and accordingly remains impervious to digestion (Zhou et al., 2010).
The structural or compositional nature of a food can sufficiently restrict the amylolysis of the starch component of that food material. Where small intestinal uptake of hydrolysed starch does not occur, then digestion can still take place through fermentation in the colon. This is said to occur for non-glycaemic carbohydrates, which includes the so-called “resistant” starch as well as soluble polysaccharide components (such as pectin), and insoluble fibres (e.g. cell wall materials including cellulose and hemi-celluloses. The digestion of these materials will be discussed in the next section.
5.2 Digestion of Non-glycaemic Carbohydrates
5.2.1 Resistant Starch
Native starch in plant materials exists in the form of granules in which the starch is arranged in the semi-crystalline state. Granule size and shape can vary depending on the source material ranging from approximately 1–100 μm in size. The amylose fraction of starch tends to form single helical structures that are able to align into a double stranded crystallite arrangement whilst amylopectin is considered as forming double helix arrangements which can undergo further self-assembly into radially expanding domains of crystalline and amorphous structures. The crystalline structures, as present in the native plant source are effectively resistant to enzymatic hydrolysis (providing the terminology “resistant starch”), with amylose considered to be the more resistant of the two fractions (Sajilata, Singhal, & Kulkarni, 2006).
As indicated, heating in the presence of water leads to the formation of glycaemic starch due to expansion and hydration of the amylose and amylopectin structures. This can occur during the cooking of raw materials, such as rice and potato, as well as during the thermal processing of manufactured foods for which derivatised starch is used as an ingredient. Amylose tends to form stronger crystalline assemblies that are surprisingly hydrophobic, and thus requires higher temperatures to undergo gelatinisation (Gallant, Bouchet, Buleon, & Perez, 1992). Accordingly, high amylose starches such as maize (or starch compositions with amylose added) can show resistant behaviours when the thermal processing is insufficient to disrupt the native crystalline state. This can also occur when starches are heated in the absence of water, which can be inhibitory to pasting. Under such circumstances, the starch component of low moisture baked or extruded foods can display delayed digestive behaviours. It should be noted that whilst maintaining the crystalline state of the native starch structure during processing is able to impart resistant behaviours during digestion, this can invariably lead to a loss of technical functionality, as the starch will not undergo pasting and gelatinisation. In this regard, product manufacturers need to have clarity about the rationale for starch inclusion as part of formulation design.
Another mechanism by which starch can demonstrate resistant properties is through retrogradation, which is a time-dependent recrystallization phenomenon that can occur after gelation has taken place—notably for high amylose starch compositions. This is the effect that causes the staling of bread, and can result in formation of increasing amounts of retrograded resistant starch (noting that for bread, such an effect is invariably considered a loss of product quality). Retrogradation can also lead to increased levels of resistant starch in cooked rice, in which cooled storage of the rice after cooking can cause recrystallization. This is exemplified in research undertaken by Nakayoshi and co-workers (Nakayoshi et al., 2015) determined that levels (% dry weight basis) of resistant starch increased from 2.5% for cooked rice to 7% when the rice was stored overnight at 4 °C. In the case of rice flour, the resistant starch content increased from 1.5% (after cooking) to 5% after chilled storage. These effects were most apparent in high amylose rice cultivars, noting that low amylose cultivars did not demonstrate any significant increase in resistant starch levels on storage (Hu, Zhao, Duan, Zhang, & Wu, 2004). Manufactured foods comprising starches used for the purposes of texturisation tend to be formulated with modified starches. Chemical reaction enables the attachment of various side groups (e.g. phosphorylation and hydroxylation) to the starch chains, which can provide a broader range of material and function properties. One particular consequence of the introduction of these side groups is that modified starches display a greater resistance to retrogradation than native starches, noting that such modifications do not appear to greatly influence the susceptibility of starch molecules towards amylolysis.
The relative balance of glycaemic and non-glycaemic starch has consequences for the levels of blood glucose arising from digestion. Purely glycaemic starch when consumed in foods that readily undergo rapid digestive breakdown, can lead to elevated postprandial blood sugar levels, are fall under the definition of high glycaemic index (GI) foods. The GI of glycaemic starches can be reduced by the matrix properties of the food, in which slower disintegration or more densely structured digesta can reduce the rate of starch hydrolysis and uptake (often termed as slowly digestible starch). In the case of resistant starch, the greater imperviousness to hydrolysis leads to a marked decrease in blood sugar levels and accordingly, foods comprising higher concentrations of resistant starch tend to be categorised as being of low GI. The lower GI values associated with resistant starch are considered as beneficial in relation to a number of physiological biomarkers. The suppression of postprandial glucose is understood to improve insulin response as well as promoting lipid oxidation. Health benefits are reported to support mitigating obesity and reducing the propensity towards type 2 diabetes (Ashwar, Gani, Shah, Wani, & Masoodi, 2016).
Whilst resistant starch is less effectively digested in the small intestine, it is able to be broken down into utilisable by-products in the large intestine. The colonic microflora is able to ferment resistant starch fractions producing a number of short chain fatty acids. Short chain fatty acid profiling has indicated high levels of butyrate and somewhat lower levels of acetate being produced in comparison with other forms of non-glycaemic edible fibres. Short chain fatty acids provide an effective energy source for the colonic cells as well as broader utilisation by the body. Human studies have indicated that between 30 and 70% of resistant starch is metalisable, with the remainder excreted in faeces (Cummings, Beatty, Kingman, Bingham, & Englyst, 1996; Ranhotra, Gelroth, & Glaser, 1996). Variations in levels of malabsorbed resistant starch tend to arise as a consequence of the amount of starch consumed in the study design.
5.2.2 Non-starch Soluble and Insoluble Dietary Fibres
In addition to resistant starch a number of non-glycaemic carbohydrate material are consumed as part of regular dietary food intake (Dhingra, Michael, Rajput, & Patil, 2012). The generic terminology of these is fibre and this term can be further segmented into soluble and insoluble fibre. Both types are defined as being resistant to digestion and absorption in the small intestine, but capable of undergoing partial or complete fermentation in the large intestine. Both soluble and insoluble fibres are regularly encountered as part of consumption of fruit and vegetable and cereals. Insoluble fibre tends to be derived from the main structural elements in plant systems, notably as components in cell walls, such as cellulose and hemicellulose, or as a reinforcing component between cell wall structures as in the case of lignin. Whilst cellulose is a hydrophilic biopolymer possessing a linear primary structure comprising multiple glucose monosaccharides. In the absence of any branching or secondary or tertiary structure, its rigid, rod-like conformation allow it to undergo extensive intermolecular hydrogen bonding leading to the formation of fibrillary crystallites that are insoluble and, unlike starch, does not or hydrate on heating in aqueous media. In humans, it is also completely resistant to hydrolysis by any digestive enzymes within the gastrointestinal tract. Lignin is equally insoluble in water, but is markedly more hydrophobic than cellulose, possessing a highly branched phenolic structure. As with cellulose, it is impervious to gastrointestinal hydrolysis or digestion.
Soluble fibres are also inherently present in dietary fruit and vegetables and are generally classed as polysaccharides. Chief amongst these are the pectins, but also other soluble biopolymers such as the fructans (a class of polysaccharide that includes inulin and which comprises fructose units), and the glucans. Pectin itself also contributes to cell wall mechanics, notably in supporting cell wall extension during plant growth. Whilst pectins can show some variation in molecular structure and composition, they share a common linear polymer backbone based on interspersed galacturonic acid and galacturonic acid methyl ester units in an interrupted repeat arrangement. Compositional variations are due to side chain attachment by various saccharide units. The functional properties of pectins vary accordingly with composition, demonstrating the ability to provide viscosification and gelling behaviours depending on conditions. All forms of pectin are resistant to small intestinal digestion but can undergo varying degrees of fermentation in the large intestine.
Ingestion of natural insoluble and soluble fibre is recognised as having a number of dietary and health related impacts (Potty, 1996; Slavin, 2013). The release of short chain fatty acids during colonic fermentation provide an energy source for the microbiota, as well as undergoing absorption in the colon. Dietary fibre may additionally assist in the regulation of blood sugar levels, as well as reducing total and LDL cholesterol levels. The effect of water binding and viscosification of both soluble and insoluble fibre may also to reduce rate of gastric emptying thereby enhancing satiety signalling, as well facilitating gastrointestinal transit and aiding in faecal bulking.
Consumption of fruit and vegetable rich foods provide one pathway for incorporation of fibre as part of dietary intake. However, manufactured foods may contain varying degrees of both soluble and insoluble fibres as part of product formulation. These fibre components are isolated from diverse raw materials and (particularly in the case of soluble polysaccharides) provide a broader spectrum of dietary polysaccharides than would normally be present in wholefoods. Thus, in addition to pectin (commercially derived various fruit sources such as citrus peel and apple pomace), other polysaccharides such as the alginates and carrageenans (from seaweed), guar and locust bean gum (from seeds), gum arabic and gum tragacanth (exudate gums) and xanthan (microbial expression) are now widely used in food production. These various polysaccharides are extracted from source materials for which their native function ranges from structural support through to acting as an energy source, or (as in the case of the exudate gums) providing a wound healing mechanism against structural damage.
The main soluble fibres materials used in food manufacturing share some common attributes, namely being high molecular weight biopolymer species assembled as extended chains of monosaccharide units. Variation in primary structure occurs through the presence and location of the different monosaccharide units comprising the polymer chain, whether these are non-ionic or ionic (noting that water structuring capacity of ionic polysaccharides can be sensitive to variable pH and ionic conditions), and whether the chains are linear or branched. The conformation of the primary structure can influence intramolecular interactions (predominantly hydrogen bonding) within the chain, leading to varying degrees of folding. The nature of polysaccharide secondary and tertiary structure can in turn impact on the ability of polysaccharides to undergo intermolecular interaction, leading to the formation of quaternary network assemblies, and which is characteristic of gelling polysaccharides. In some cases, such as for the alginates and cellulose derivatives, additional chemical or biochemical treatments are able to alter the native molecular structure as a means of further tailoring physicochemical properties. Accordingly, their function in foods is primarily technical, enabling control over material properties through viscosification and/or gelation as well as other abilities such as emulsification, inhibition of ice recrystallization, film forming and foaming. For high moisture products such as ice creams, dressings and sauces, inclusion of soluble polysaccharides can greatly enhance product attributes, such as sensory properties, physical stability and shelf life extension.
In this context, whilst these materials can be classified as dietary fibres, they are not typically used for nutritional supplementation in manufactured food systems. This is primarily due to the fact that that the dosage levels to achieve a particular technical effect within a product are usually a few tenths of a per cent of the combined formulation, and thus not considered as being of sufficiently high concentration to achieve a nutritional benefit. That said, there has been increasing interest in recent years as to the function and role of isolated insoluble and soluble fibres during digestion (Lovegrove et al., 2017; Noack, Timm, Hospattankar, & Slavin, 2013). The inability to be digested in the stomach and small intestine and retention of native structure can enable the water structuring characteristics of polysaccharides to be at least partially retained during the digestion process. As discussed earlier, gastrointestinal motility can be manipulated in relation to the material properties of the digesta. Viscous or gelled materials can slow down the rate of gastric emptying and increase transit time within the small intestine, positively influencing satiation. The viscous properties of digesta can also serve to influence the rate of diffusion of nutrients to the epithelium. Such effects have been explored for a number of different polysaccharide systems. The digestive properties of sodium alginate, for example, have come under particular scrutiny for its ability to enable varying structural states in the stomach and small intestine, ranging from varying degrees of viscosity through to formation of gelled structures, for example, in the presence of calcium (Hoad et al., 2004) or (under acidic conditions in the stomach) through synergistic interactions with other polysaccharides such as pectin.
These effects have been mainly considered for pharmaceutical applications for which the polysaccharide is the primary structuring material, allowing the structural and material properties to be highly defined, as well as potentially bypassing consideration such as oral processing. Specific functions have included the controlled delivery of drugs, and a particularly novel application as a gastric raft (i.e. forming a structured environment of the surface layer of the gastric fluid), that has been demonstrated as being efficacious at mitigating the symptoms of acid reflux (Jang, Lee, Ryu, Son, & Kang, 2014). Other physiological effects, such as manipulating satiety have also been extensively studied for alginate as well as other polysaccharide systems, and whilst the use of model systems has demonstrated that effects can be generated that are influenced by the material properties of the polysaccharide during digestion, challenges remain in translating these effects into actual food systems, where the maintenance of the target structure through oral processing, gastric and small intestinal residence can be difficult to achieve, whilst still retaining the expected eating properties of the food in question.
Research interest in the digestive behaviours of polysaccharide carbohydrates has also tended to focus on those systems already permitted for use in food or pharmaceuticals. This is partly due to the challenges of bringing new materials through the clearance processes required for utilisation in food manufacturing. However, it is also the case that there are still countless materials naturally present in flora and fauna that may have specific functional digestive behaviours that could be of utilisation in both food and drug systems. One recent particular example relates to the potential use of a native New Zealand gum in reducing food intake (Wee, Lentle, Goh, & Matia-Merino, 2017). The gum in question is extracted from the mamaku black tree fern. The polysaccharide composition within the fern is quite complex, being predominantly a glucuronomannan comprising a backbone of 4-linked methylesterified glucopyranosyl uronic acid and 2-linked mannopyranosyl residues, branched at 0.3 of 45% and at both 0.3 and 0.4 of 53% of the mannopyranosyl residues with side chains likely comprising terminal xylopyranosyl, terminal galactopyranosyl, non-methylesterified terminal glucopyranosyl uronic acid and 3-linked glucopyranosyl uronic acid residues (Wee, Matia-Merino, Carnachan, Sims, & Goh, 2014).
This polysaccharide shows some unusual rheological behaviour when isolated and solubilised. Dependent on concentration and ionic environment, it behaves as a pseudoplastic fluid at shear rates typically greater than 4–10 s−1. However, before onset of shear thinning behaviour, there is an intermediate shear rate region for which the polysaccharide demonstrates rheopectic properties. This is unusual behaviour for a polysaccharide system, and represents a material that can potentially undergo shear thickening under the shear conditions present in the stomach. As part of an in vivo study, Wee and co-authors investigated the food intake of rats that had been gavaged with a solution of the gum and evidenced a reduction in extent of gastric emptying with accompanying suppression of appetite and food intake relative to a control group that had not been gavaged with the gum solution. Whilst these findings demonstrated a clear impact of the gum on eating behaviour, there are acknowledged challenges associated with the progression of these effects into food systems for human consumption with comparable physiological outcomes. Nevertheless, these findings not only support prior research highlighting the impact of water structuring on digestive behaviours but also demonstrate the opportunities arising from exploration of materials outside of the scope of those currently used in food manufacture.
- Akkermans, C., Van Der Goot, A. J., Venema, P., Gruppen, H., Vereijken, J. M., Van Der Linden, E., et al. (2007). Micrometer-sized fibrillar protein aggregates from soy glycinin and soy protein isolate. Journal of Agricultural and Food Chemistry, 55(24), 9877–9882. https://doi.org/10.1021/jf0718897CrossRefPubMedGoogle Scholar
- Arancibia, C., Jublot, L., Costell, E., & Bayarri, S. (2011). Flavor release and sensory characteristics of o/w emulsions. Influence of composition, microstructure and rheological behavior. Food Research International, 44(6), 1632–1641. https://doi.org/10.1016/j.foodres.2011.04.049CrossRefGoogle Scholar
- Barbe, F., Menard, O., Le Gouar, Y., Buffiere, C., Famelart, M. H., Laroche, B., et al. (2014). Acid and rennet gels exhibit strong differences in the kinetics of milk protein digestion and amino acid bioavailability. Food Chemistry, 143, 1–8. https://doi.org/10.1016/j.foodchem.2013.07.100CrossRefPubMedPubMedCentralGoogle Scholar
- Bellesi, F. A., Martinez, M. J., Ruiz-Henestrosa, V. M. P., & Pilosof, A. M. R. (2016). Comparative behavior of protein or polysaccharide stabilized emulsion under in vitro gastrointestinal conditions. Food Hydrocolloids, 52, 47–56. https://doi.org/10.1016/j.foodhyd.2015.06.007CrossRefGoogle Scholar
- Chen, B. C., McClements, D. J., & Decker, E. A. (2013). Design of foods with bioactive lipids for improved health. In M. P. Doyle & T. R. Klaenhammer (Eds.), Annual review of food science and technology (Vol. 4, pp. 35–56). Palo Alto: Annual Reviews.Google Scholar
- Ellrichmann, M., Kapelle, M., Ritter, P. R., Holst, J. J., Herzig, K. H., Schmidt, W. E., et al. (2008). Orlistat inhibition of intestinal lipase acutely increases appetite and attenuates postprandial glucagon-like peptide-1-(7-36)-amide-1, cholecystokinin, and peptide YY concentrations. Journal of Clinical Endocrinology and Metabolism, 93(10), 3995–3998. https://doi.org/10.1210/jc.2008-0924CrossRefPubMedGoogle Scholar
- Ferrer, E., Alegria, A., Farre, R., Abellan, P., Romero, F., & Clemente, G. (2003). Evolution of available lysine and furosine contents in milk-based infant formulas throughout the shelf-life storage period. Journal of the Science of Food and Agriculture, 83(5), 465–472. https://doi.org/10.1002/jsfa.1402CrossRefGoogle Scholar
- Foltz, M., Maijaars, J., Schuring, E. A. H., van der Wal, R. J. P., Boer, T., Duchateau, G. S. M., et al. (2009). Intragastric layering of lipids delays lipid absorption and increases plasma CCK but has minor effects on gastric emptying and appetite. American Journal of Physiology—Gastrointestinal and Liver Physiology, 296(5), G982–G991. https://doi.org/10.1152/ajpgi.90579.2008CrossRefPubMedGoogle Scholar
- Grundy, M. M. L., Grassby, T., Mandalari, G., Waldron, K. W., Butterworth, P. J., Berry, S. E. E., et al. (2015). Effect of mastication on lipid bioaccessibility of almonds in a randomized human study and its implications for digestion kinetics, metabolizable energy, and postprandial lipemia. American Journal of Clinical Nutrition, 101(1), 25–33. https://doi.org/10.3945/ajcn.114.088328CrossRefPubMedGoogle Scholar
- Hooton, D., Lentle, R., Monro, J., Wickham, M., & Simpson, R. (2015). The secretion and action of brush border enzymes in the mammalian small intestine. In B. Nilius, T. Gudermann, R. Jahn, R. Lill, O. H. Petersen, & P. P. DeTombe (Eds.), Reviews of physiology, biochemistry and pharmacology (Vol. 168, pp. 59–118). Switzerland: Springer.CrossRefGoogle Scholar
- Jang, S. W., Lee, J. W., Ryu, D. S., Son, M., & Kang, M. J. (2014). Design of pH-responsive alginate raft formulation of risedronate for reduced esophageal irritation. International Journal of Biological Macromolecules, 70, 174–178. https://doi.org/10.1016/j.ijbiomac.2014.06.048CrossRefPubMedGoogle Scholar
- Kelly, S. M., Shorthouse, M., Cotterell, J. C., Riordan, A. M., Lee, A. J., Thurnham, D. I., et al. (1998). A 3-month, double-blind, controlled trial of feeding with sucrose polyester in human volunteers. British Journal of Nutrition, 80(1), 41–49. https://doi.org/10.1017/s0007114598001755CrossRefPubMedGoogle Scholar
- Keogh, J. B., Wooster, T. J., Golding, M., Day, L., Otto, B., & Clifton, P. M. (2011). Slowly and rapidly digested fat emulsions are equally satiating but their triglycerides are differentially absorbed and metabolized in humans. Journal of Nutrition, 141(5), 809–815. https://doi.org/10.3945/jn.110.131110CrossRefPubMedGoogle Scholar
- Lasse, M., Ulluwishewa, D., Healy, J., Thompson, D., Miller, A., Roy, N., et al. (2016). Evaluation of protease resistance and toxicity of amyloid-like food fibrils from whey, soy, kidney bean, and egg white. Food Chemistry, 192, 491–498. https://doi.org/10.1016/j.foodchem.2015.07.044CrossRefPubMedGoogle Scholar
- Lueamsaisuk, C., Lentle, R. G., MacGibbon, A. K. H., Matia-Merino, L., & Golding, M. (2014). Factors influencing the dynamics of emulsion structure during neonatal gastric digestion in an in vitro model. Food Hydrocolloids, 36, 162–172. https://doi.org/10.1016/j.foodhyd.2013.09.013CrossRefGoogle Scholar
- Lueamsaisuk, C., Lentle, R. G., MacGibbon, A. K. H., Matia-Merino, L., & Golding, M. (2015). The effect of lactoferrin on physical changes in phospholipid stabilised emulsions during neonatal in vitro gastric digestion: Does synergism of pepsin and lipase promote lipolysis in protein-stabilised emulsions? Food Hydrocolloids, 43, 785–793. https://doi.org/10.1016/j.foodhyd.2014.08.010CrossRefGoogle Scholar
- Lundin, L., & Golding, M. (2009). Structure design for healthy food. Australian Journal of Dairy Technology, 64(1), 68–74.Google Scholar
- Mackie, A. R., Rafiee, H., Malcolm, P., Salt, L., & van Aken, G. (2013). Specific food structures supress appetite through reduced gastric emptying rate. American Journal of Physiology-Gastrointestinal and Liver Physiology, 304(11), G1038–G1043. https://doi.org/10.1152/ajpgi.00060.2013CrossRefPubMedPubMedCentralGoogle Scholar
- Maldonado-Valderrama, J., Gunning, A. P., Ridout, M. J., Wilde, P. J., & Morris, V. J. (2009). The effect of physiological conditions on the surface structure of proteins: Setting the scene for human digestion of emulsions. European Physical Journal E, 30(2), 165–174. https://doi.org/10.1140/epje/i2008-10426-0CrossRefGoogle Scholar
- Mandalari, G., Grundy, M. M. L., Grassby, T., Parker, M. L., Cross, K. L., Chessa, S., et al. (2014). The effects of processing and mastication on almond lipid bioaccessibility using novel methods of in vitro digestion modelling and micro-structural analysis. British Journal of Nutrition, 112(9), 1521–1529. https://doi.org/10.1017/s0007114514002414CrossRefPubMedGoogle Scholar
- Marciani, L., Faulks, R., Wickham, M., Bush, D., Pick, B., Wright, J., et al. (2009). Effect of intragastric acid stability of fat emulsions on gastric emptying, plasma lipid profile and postprandial satiety. British Journal of Nutrition, 101(6), 919–928. https://doi.org/10.1017/S0007114508039986CrossRefPubMedGoogle Scholar
- Marciani, L., Wickham, M. S. J., Bush, D., Faulks, R., Wright, J., Fillery-Travis, A., et al. (2006). Magnetic resonance imaging of the behaviour of oil-in-water emulsions in the gastric lumen of man. British Journal of Nutrition, 95(2), 331–339. https://doi.org/10.1079/BJN20051628CrossRefPubMedGoogle Scholar
- Marciani, L., Wickham, M., Singh, G., Bush, D., Pick, B., Cox, E., et al. (2006). Delaying gastric emptying and enhancing cholecystokinin release and satiety by using acid stable fat emulsions. Gastroenterology, 130(4), A227–A227.Google Scholar
- Marciani, L., Wickham, M., Singh, G., Bush, D., Pick, B., Cox, E., et al. (2007). Enhancement of intragastric acid stability of a fat emulsion meal delays gastric emptying and increases cholecystokinin release and gallbladder contraction. American Journal of Physiology-Gastrointestinal and Liver Physiology, 292(6), G1607–G1613. https://doi.org/10.1152/ajpgi.00452.2006CrossRefPubMedGoogle Scholar
- Mullally, M. M., Mehra, R., & FitzGerald, R. J. (1998). Thermal effects on the conformation and susceptibility of beta-lactoglobulin to hydrolysis by gastric and pancreatic endoproteinases. Irish Journal of Agricultural and Food Research, 37(1), 51–60.Google Scholar
- Mun, S., Decker, E. A., Park, Y., Weiss, J., & McClements, D. J. (2006). Influence of interfacial composition on in vitro digestibility of emulsified lipids: Potential mechanism for chitosan’s ability to inhibit fat digestion. Food Biophysics, 1(1), 21–29. https://doi.org/10.1007/s11483-005-9001-0CrossRefGoogle Scholar
- Nakayoshi, Y., Nakamura, S., Kameo, Y., Shiiba, D., Katsuragi, Y., & Ohtsubo, K. (2015). Measurement of resistant starch content in cooked rice and analysis of gelatinization and retrogradation characteristics. Bioscience Biotechnology and Biochemistry, 79(11), 1860–1866. https://doi.org/10.1080/09168451.2015.1044934CrossRefGoogle Scholar
- Nik, A. M., Wright, A. J., & Corredig, M. (2011). Impact of interfacial composition on emulsion digestion and rate of lipid hydrolysis using different in vitro digestion models. Colloids and Surfaces B-Biointerfaces, 83(2), 321–330. https://doi.org/10.1016/j.colsurfb.2010.12.001CrossRefGoogle Scholar
- Noack, J., Timm, D., Hospattankar, A., & Slavin, J. (2013). Fermentation profiles of wheat dextrin, inulin and partially hydrolyzed guar gum using an in vitro digestion pretreatment and in vitro batch fermentation system model. Nutrients, 5(5), 1500–1510. https://doi.org/10.3390/nu5051500CrossRefPubMedPubMedCentralGoogle Scholar
- Norton, J. E., Wallis, G. A., Spyropoulos, F., Lillford, P. J., & Norton, I. T. (2014). Designing food structures for nutrition and health benefits. In M. P. Doyle & T. R. Klaenhammer (Eds.), Annual review of food science and technology (Vol. 5, pp. 177–195). Palo Alto: Annual Reviews.Google Scholar
- Potty, V. H. (1996). Physio-chemical aspects, physiological functions, nutritional importance and technological significance of dietary fibres—A critical appraisal. Journal of Food Science and Technology-Mysore, 33(1), 1–18.Google Scholar
- Ranhotra, G. S., Gelroth, J. A., & Glaser, B. K. (1996). Energy value of resistant starch. Journal of Food Science, 61(2), 453–455. https://doi.org/10.1111/j.1365-2621.1996.tb14215.xCrossRefGoogle Scholar
- Roos, N., Lorenzen, P. C., Sick, H., Schrezenmeir, J., & Schlimme, E. (2003). Cross-linking by transglutaminase changes neither the in vitro proteolysis nor the in vivo digestibility of caseinate. Kieler Milchwirtschaftliche Forschungsberichte, 55(4), 261–276.Google Scholar
- Rui, X., Fu, Y. T., Zhang, Q. Q., Li, W., Zare, F., Chen, X. H., et al. (2016). A comparison study of bioaccessibility of soy protein gel induced by magnesium chloride, glucono-delta-lactone and microbial transglutaminase. Lwt-Food Science and Technology, 71, 234–242. https://doi.org/10.1016/j.lwt.2016.03.032CrossRefGoogle Scholar
- Sajilata, M. G., Singhal, R. S., & Kulkarni, P. R. (2006). Resistant starch—A review. Comprehensive Reviews in Food Science and Food Safety, 5(1), 1–17. https://doi.org/10.1111/j.1541-4337.2006.tb00076.xCrossRefGoogle Scholar
- Sarkar, A., Goh, K. K. T., & Singh, H. (2010a). Properties of oil-in-water emulsions stabilized by beta-lactoglobulin in simulated gastric fluid as influenced by ionic strength and presence of mucin. Food Hydrocolloids, 24(5), 534–541. https://doi.org/10.1016/j.foodhyd.2009.12.005CrossRefGoogle Scholar
- Schmidt, D. G., & Vanmarkwijk, B. W. (1993). Enzymatic-hydrolysis of whey proteins—Influence of heat-treatment of alpha-lactalbumin and beta-lactoglobulin on their proteolysis by pepsin and papain. Netherlands Milk and Dairy Journal, 47(1), 15–22.Google Scholar
- Seimon, R. V., Wooster, T., Otto, B., Golding, M., Day, L., Little, T. J., et al. (2009). The droplet size of intraduodenal fat emulsions influences antropyloroduodenal motility, hormone release and appetite in healthy males. American Journal of Clinical Nutrition, 89(6), 1729–1736. https://doi.org/10.3945/ajcn.2009.27518CrossRefPubMedGoogle Scholar
- van Aken, G. A., Bomhof, E., Zoet, F. D., Verbeek, M., & Oosterveld, A. (2011). Differences in in vitro gastric behaviour between homogenized milk and emulsions stabilised by Tween 80, whey protein, or whey protein and caseinate. Food Hydrocolloids, 25(4), 781–788. https://doi.org/10.1016/j.foodhyd.2010.09.016CrossRefGoogle Scholar
- Venema, K., Minekus, A., & Havenaar, R. (2004). Advanced in-vitro models of the gastro-intestinal tract—Novel tools to study functionality of dietary fibres. In J. W. VanderKamp, N. G. Asp, J. M. Jones, & G. Schaafsma (Eds.), Dietary fibre—Bio-active carbohydrates for food and feed (pp. 99–112). Wageningen: Academic Publishers.Google Scholar
- Wee, M. S. M., Matia-Merino, L., Carnachan, S. M., Sims, I. M., & Goh, K. K. T. (2014). Structure of a shear-thickening polysaccharide extracted from the New Zealand black tree fern, Cyathea medullaris. International Journal of Biological Macromolecules, 70, 86–91. https://doi.org/10.1016/j.ijbiomac.2014.06.032CrossRefPubMedGoogle Scholar