Search

Food structure affects the breakdown kinetics of food. This has been extensively studied via in vitro digestion models. However, these in vitro findings need to be verified in vivo. Previously, we found that MRI parameters (T1 and T2 relaxation times) can serve as markers of protein digestion. Here, we assessed the effect of food hardness and protein content on gastric emptying as well as the application of these MRI parameters to monitor in vivo gastric digestion of solid foods. In a randomized cross-over trial, 18 healthy males (age: 27 ± 5 y; BMI 23 ± 1 kg/m2) ingested three gels: a soft gel with low protein content (Soft-LP), a hard gel with low protein content (Hard-LP), and a hard gel with high protein content (Hard-HP). Before and every 10 min after ingestion till 85 min, abdominal MRI scans were obtained to determine the gastric content volume and T1 and T2 of the gastric content. The decrease in gastric content volume differed among gels: Hard-HP < Hard-LP < Soft-LP. For all gels, mean T1 and T2 of the measured stomach content decreased directly after ingestion and then after 15 min gradually increased with Hard-HP < Hard-LP < Soft-LP. In conclusion, high protein content and high hardness slowed gastric emptying down. In addition, we show that T1 and T2 measurements provide extra information on the dilution and digestion of solid foods in the stomach. This underpins the potential of MRI to provide insights into in vivo protein digestion and link in vitro and in vivo digestion research.

Enzyme-catalysed hydrolysis is important in protein digestion. Protein hydrolysis is initiated by pepsin at low pH in the stomach. However, pepsin action and acidification happen simultaneously to gastric emptying, especially for liquid meals. Therefore, different extents of exposure to the gastric environment change the composition of the chyme that is emptied from the stomach into the small intestine over time. We assessed the susceptibility of a protein to trypsin-catalysed hydrolysis in the small intestine, depending on its pH and hydrolysis history, simulating chyme at different times after the onset of gastric emptying. Isothermal titration calorimetry was used to study the kinetics of pepsin and trypsin-catalysed hydrolysis. Bovine serum albumin (BSA) that was acidified and hydrolysed with pepsin, showed the highest extent and most efficient hydrolysis by trypsin. BSA in the chyme that would be first emptied from the stomach, virtually bypassing gastric acidity and peptic action, reduced trypsin-catalysed hydrolysis by up to 58% compared to the acidified, intact protein, and 77% less than the acidified, pepsin-hydrolysate. The least efficient substrate for trypsin-catalysed hydrolysis was the acidified, intact protein with a specificity constant (kcat/Km) nearly five times lower than that of the acidified, pepsin-hydrolysate. Our results illustrate the synergy between pepsin and trypsin hydrolysis, and indicate that gastric hydrolysis increases the efficiency of the subsequent trypsin-catalysed hydrolysis of a model protein in the small intestine. This journal is

Process streams of agro‐food industries are often large and viscous. In order to fractionate such a stream the viscosity can be reduced by either a high temperature or dilution, the former is not an option in case of temperature sensitive components. Such streams are diluted prior to chromatographic fractionation, resulting in even larger volumes and high energy costs for sub‐sequential water removal. The influence of feed viscosity on the performance of simulated moving bed chromatography has been investigated in a case study of the recovery of a γ‐aminobutyric acid rich fraction from tomato serum. This work addresses the chromatographic system design, evaluates results from a pilot scale operation, and uses these to calculate the productivity and water use at elevated feed concentration. At the two higher feed viscosities (2.5 and 4 mPa·s) water use is lower and productivity higher, compared to the lowest feed viscosity (1 mPa·s). The behavior of the sugars for different feed viscosities can be described well by the model using the ratio of feed to eluent as dilution factor. The behavior of γ‐aminobutyric acid is highly concentration dependent and the recovery could not be accurately predicted.

In this paper, we present the rich physics and chemistry of the gastric digestion of protein gels. Knowledge of this matter is important for the development of sustainable protein foods that are based on novel proteins sources like plant proteins or insects. Their digestibility is an important question in the design of these new protein foods. As polyelectrolyte gels, they can undergo volume changes upon shifts in pH or ionic strengths, as protein gels experience when entering the gastric environment. We show that these volume changes can be modelled using the Flory-Rehner theory, combined with Gibbs-Donnan theory accounting for the distribution of electrolytes over gel and bath. In-vitro experiments of soy protein gels in simulated gastric fluid indeed show intricate swelling behaviour, at first the gels show swelling but at longer times they shrink again. Simulations performed with the Flory-Rehner/Gibbs-Donnan theory reproduce qualitatively similar behaviour. In the final part of the paper, we discuss how the model must be extended to model realistic conditions existing in the in-vivo gastric environment., Highlights • Physical-chemistry of gastric digestion of protein gels is presented. • A numerical model based on Flory-Rehner can qualitatively explain swelling of gels. • Protein gels in gastric environments can swell and shrink when put in gastric acid. • For ion transport it is important to recognize the finite volume of the gastric juice.

We report on the effect of processing, particularly heating, on the digestion dynamics of pea proteins using the standardised semi-dynamic in vitro digestion method. Fractions with native proteins were obtained by mild aqueous fractionation of pea flour. A commercial pea protein isolate was chosen as a benchmark. Heating dispersions of pea flour and mild protein fractions reduced the trypsin inhibitory activity to levels similar to that of the protein isolate. Protein-rich and non-soluble protein fractions were up to 18% better hydrolysed after being thermally denatured, particularly for proteins emptied later in the gastric phase. The degree of hydrolysis throughout the digestion was similar for these heated fractions and the conventional isolate. Further heating of the protein isolate reduced its digestibility as much as 9%. Protein solubility enhances the digestibility of native proteins, while heating aggregates the proteins, which ultimately reduces the achieved extent of hydrolysis from gastro-small intestinal enzymes.

We present a dynamic, semi-mechanistic, compartmental protein digestion model to study the kinetics of protein digestion. The digestive system is described as a series of eight compartments: one for the stomach, one for the duodenum, two for the jejunum and four for the ileum. The digestive processes are described by a set of zero or first order differential equations. The model considers ingestion of a meal, secretion of gastric and pancreatic juices, protein hydrolysis, grinding, transit and amino acid absorption. The model was used to simulate protein digestion of a meal composed of a solid and a liquid phase or one where both phases are blended into a homogeneous phase. Luminal volumes and pH of gastric and duodenal contents were estimated for both meals. Further, gastric emptying is described as a function of the energy density of the bolus, instead of the more common mass action approach.

International audience; Background: In vitro digestion methods are widely used to investigate the effect of food properties on the hydrolysis of the main macronutrients: starch, lipid and protein. The growing quantity of experimental data calls for strategies to quantitatively compare the effect of food composition and structure on their hydrolysis kinetics. Mathematical modelling is a powerful tool for this purpose as it allows to summarize complex phenomena into a few equations, and quantify relevant model parameters. Scope and approach: This review focuses on modelling in vitro digestion data, more particularly the hydrolysis of the main macronutrients at the gastric and small intestinal stages. Both static and dynamic in vitro conditions are considered, giving an overview of the modelling strategies available for each macronutrient. Besides, ongoing efforts to model the effects of food micro-and macrostructure as well as the interplay between macronutrient hydrolysis are summarized. A view on how modelling may help to bridge the gap between in vitro and in vivo studies is also provided. Key findings and conclusions: In vitro digestion and mathematical modelling are highly complementary methods. Mathematical models can provide a full and quantitative picture of the phenomena taking place, meanwhile in vitro experiments offer an excellent framework to test modelling concepts and assumptions. Some hybrid strategies, combining in vitro and in silico approaches have also been proposed to more accurately translate in vitro observations into in vivo predictions. Although very young, this field of research appears very promising to complement, or offer an alternative to experimental studies.

The effects of high concentration in the fine ultrafiltration of a solution of oligosaccharides were investigated both experimentally and using a mass transfer model based on the Maxwell-Stefan equations. At high concentrations, negative retentions were found for the smaller sugars, which cannot be ascribed to effects of ionic interaction, membrane adsorption or fouling. Instead, the behaviour could be quantitatively described by incorporating the effects of the thermodynamic non-ideality of the solutions and the effects of the pore size distribution. Experiments were performed to validate the model using as feed an oligosaccharide mixture with a concentration up to a 35% w/w. The model predictions allows the identification of an optimum feed concentration at which the efficiency of the separation is maximized. The results show that the fine ultrafiltration of sugars can be well described and predicted when taking into account the relevant thermodynamic interactions, the membrane pore size distribution and pressure effects.

Despite the vast number of studies on the understanding and estimation of the permeate flux in ultrafiltration, most of them base their estimations on either one or another mechanism, without pointing out a clear ‘bridge’ between them. The aim of this paper is to assess these mechanisms on the determination of the permeate flux, using as feed a multicomponent mixture of BSA, NaCl and H2O. Maxwell-Stefan Equations expressed as function of the components' volume fractions were used for an easier consideration of the non-idealities of the system. These non-idealities (hydration, adsorption, electrical interactions and volume exclusion) were critical in the local fluxes calculation, for which an increase in the thickness of the boundary layer along the filtration channel was considered. The developed model proved to be suitable for the estimation of fluxes lower than the limiting flux. Since the non-idealities of the system can be calculated along the concentration polarization layer, no extra information on the protein diffusivity was needed. Additionally, the fact that the model includes all the components from the solution offers the possibility of including the rejection of the accompanying ions in the calculations.

Information about the sizes of the solute molecules and membrane pores is needed to estimate solute rejection in filtration processes. Molecules are normally regarded as spheres, and the Stokes radius is commonly used to represent their molecular size. However, many molecules used in food and pharma processes are oligomers or polymers which are strongly elongated; therefore, considering them spherical affects the accuracy of the model predictions. We here adapt the so-called Steric Pore Model to a more realistic representation of the transfer of rigid elongated molecules into and through ultrafiltration membrane pores. To do so, sugars with different degree of polymerization were used as model molecules. They were considered to be capsule-shaped to facilitate their size estimation. In order to represent the system as accurately as possible, the effect of hydration on the sugars size was included, and the membrane pore size distribution was estimated based on rejection data. It was demonstrated that considering these molecules to be capsule-shaped instead of spherical generates better predictions over the entire rejection spectrum using a unique pore size distribution. Additionally, this capsular geometry lets us simplify the calculations, making the estimation of the rejection straightforward.

We developed a design method for an inhomogeneous membrane cascade by adopting the McCabe-Thiele method, which is long established for designing distillation columns. The stage cut value is an independent design parameter in the design procedure and thus has to be set. Within each section, the operating conditions were uniform, but both sections could be operated differently using various combinations of membranes, trans-membrane pressure, temperature and stage cut. The procedure was applied to cascaded nanofiltration for the fractionation of a mixture of fructooligosaccharides of varying molecular weight. The stage and area requirements were strongly dependent on the initial design parameter, the overall stage cut. The total area was related to the overall system cut. However, the overall system cut was dependent on the stage cuts for both sections (top and bottom). The top stage cut could be chosen, whereas the bottom stage cut needed to be calculated iteratively to match the top design at the intersection.

Inhomogeneous membrane cascade systems have been utilized to purify fructooligosaccharides (FOS). Such a process allows a different setup at every stage of the cascade. Varying the setup at every stage implies an optimization problem related to the selection of the membrane and combinations of operating conditions. This paper solves the optimization problem for an inhomogeneous 3-stage membrane cascade and uses the solution as a design guideline. The optimization problem in the 3-stage membrane cascade design has been formulated as a mixed integer, non-linear programming model and solved using the global optimization solver, BARON. By maximizing the yield repetitively with varying purity requirements, a frontier curve has been constructed. The frontier curve was mapped showing the window of operation. The map guides towards the setup that promotes higher permeation in the feed stage when we switch from high yield to high purity. On the other hand, the setup selection at the bottom stage does not show a clear switch, which indicates that the selection at this stage is less critical.

Gastric digestion is crucial for protein breakdown. Magnetic resonance techniques have a great deal of potential but remain underexplored with regard to their application in the study of food digestion via MRI-markers, such as transverse (R2) and longitudinal (R1) relaxation rates. R2 has been used to monitor gastric digestion of whey protein gels, but only in a static in vitro model. It is essential to investigate whether relaxation rates can be valid measures of digestion under dynamic circumstances. We developed a novel MRI-compatible semi-dynamic gastric simulator (MR-GAS) that includes controlled gastric secretion, emptying and mixing at body temperature. PH and protein hydrolysis were measured during protein gel digestion in the MR-GAS. R2 and R1 of the supernatant were measured by time-domain nuclear magnetic resonance (TD-NMR). The stomach chamber of the MR-GAS was also scanned with MRI to measure R2 and R1. For TD-NMR, 99% of the variance in R2 and 96% of variance in R1 could be explained as a function of protein concentration and [H+]. For MRI, the explained variances were 99% for R2 and 60% for R1. From these analysis, the obtained equations enabled the prediction of protein concentration and pH by R2 and R1. The normalised root mean squared deviation of the predictions for protein concentration were 0.15 (NMR) and 0.18 (MRI), and for pH were 0.12 (NMR) and 0.29 (MRI). In conclusion, the MR-GAS model may be used in a clinical MRI to monitor gastric digestion under in vitro dynamic circumstances, by measuring R2 and R1. These results underscore the potential of MRI to monitor nutrients hydrolysis and pH changes in future in vivo studies.

Fructose and glucose are commonly present together in mixtures and may need to be separated. Current separation methods for these isomers are complex and costly. Nanofiltration is a cost-effective method that has been widely used for separating carbohydrates of different sizes, however, it is not commonly used for such similar molecules. Here, we report the separation of fructose and glucose in a nanofiltration system in the presence of fructooligosaccharides (FOS). Experiments were performed using a pilot-scale filtration setup using a spiral wound nanofiltration membrane with molecular weight cutoff of 1 kDa. We observed three important factors that affected the separation: (1) separation of monosaccharides only occurred in the presence of FOS and became more effective when FOS dominated the solution, (2) better separation was achieved when the monosaccharides were mainly fructose, and (3) the presence of salt improved the separation only moderately. The rejection ratio (Rf/Rg) in a fructose/glucose mixture is 0.92. We reported a rejection ratio of 0.69, which was observed in a mixture of 50 g/L FOS with a fructose to glucose ratio of 4.43. The separation is hypothesized to occur due to selective transport in the FOS layer, resulting in a preferential binding towards fructose.

Gastric digestion is crucial for protein breakdown. Although it has been widely studied with in vitro models, verification in vivo remains a big challenge. Magnetic resonance imaging (MRI) has the potential to bridge this gap. Our objective was to use the transverse relaxation time (T2) and rate (R2 = T2−1) to monitor hydrolysis of protein-rich food during in vitro gastric digestion. Whey protein solution and heat-induced hydrogels were digested by means of simulated gastric fluid (SGF). Free amino groups (–NH2 groups) and protein concentration in the supernatant were measured. T2 and R2 of the digestion mixture were determined by time-domain nuclear magnetic resonance (TD-NMR) and MRI. Subsequently, relative amplitudes (TD-NMR) for different T2 values and T2 distribution (MRI) were determined. For the solution, protein concentration and T2 did not change during digestion. For the gels, water in supernatant and gel phase could be discriminated on the basis of their T2 values. During digestion, R2 of supernatant correlated positively with protein (–NH2 groups) concentration in SGF. Also, the decrease in relative amplitude of gel fraction correlated linearly with the increase of supernatant protein concentration. MRI T2-mapping showed similar associations between R2 of supernatant and protein (–NH2 groups) concentration. In conclusion, T2-measurements by TD-NMR and MRI can be used to monitor in vitro gastric digestion of whey protein gels; TD-NMR measurements contributed to interpreting the MRI data. Thus, MRI has high potential for monitoring in vivo gastric digestion and this should be further pursued.

The protein secondary structure modifications and digestibility of red abalone muscle subjected to high hydrostatic pressure (HHP) treatments (200, 300, 400, and 500 MPa for 5 min) were evaluated. The protein structure was analysed by Fourier-Transformed Infrared spectroscopy. Protein digestibility was evaluated based on the degree of hydrolysis (DH) and peptide size distributions under in vitro gastrointestinal conditions. The intermolecular β-sheet structure was disrupted at 200 MPa, compensated by the formation of the intramolecular β-sheet. At 300 and 400 MPa, the β-sheet structure can fold on itself from the interactions that stabilize the protein structure. The 310-helix structure was significantly looser at 300 MPa. Structural modifications were accompanied by β-turn formation at 300, 400, and 500 MPa. In vitro gastrointestinal digestion is improved by HHP independently of pressure level. The results suggest that high pressure improve the DH of red abalone as a consequence of β-sheet and β-turn conformations changes. Industrial relevance The seafood industry uses high hydrostatic pressure (HHP) technology to reduce undesirable sensory changes and preserve the functional and nutritional properties of compounds. The HHP experiments contributed to unravel the impact of the different level pressure on digestibility. HHP treatment can change the secondary structures of proteins and improve the protein digestibility as function the pressure level. The results of this study provide valuable information for the potential application of HHP on the development of red abalone with high-nutritional value.

The quickly expanding field of plant-based food, generally uses protein concentrates or isolates as protein source. It is however not clear to what extent the intensive processing of these raw materials affects their digestibility. We here report on the in vitro gastric digestibility of the structures present in unheated and heated dispersions of spray-dried protein isolates of soybean and yellow pea. Unheated dispersions consist primarily of insoluble individual spray-dried particles, agglomerates of these and only a small fraction of soluble protein. Pepsin activity was followed in real-time through microscopic observations, showing the disassociation of agglomerates and inward-breakdown of individual particles, which are otherwise stable at gastric pH and ionic strength. This demonstrates that solubility is not necessarily an incentive for gastric protein digestion. Heating does not significantly affect the overall digestibility of protein isolate dispersions. Nevertheless, heating disrupts the structure of spray-dried particles, increasing the amount of smaller and better digestible particles that remain suspended after centrifugation. Conversely, heat-induced aggregates remain in the pellet and are up to 50% less digestible than their unheated counterparts. This impaired digestibility is counterbalanced by a reduced proportion of poorly-digestible species in the full system (up to 11% for soy and 23% for pea).

The strength of chromatography lies in the ability of fine-tuning recovery for specific target components or fractions of interest. A downside of industrial chromatography is the need to dilute streams, as it is often applied today. This article challenges the conventional low concentration of input streams and investigates size exclusion chromatography at concentrated streams of high viscosity. Chromatographic operation with concentrated streams leads to an increased pressure drop over the column and decreased mass transfer kinetics, but also lower volumes compared to diluted streams. The objective of this research was to investigate separation performance and system dimensions as a function of viscosity for food type streams, in scenarios where viscosity is not caused by target components. Disadvantages due to increased stream volume with decreasing concentration and benefits due to decreased viscosity were evaluated, aiming to find minimal column volume. Separation performance was evaluated for a range of target components in a preparative lab-scale system using a size exclusion resin and mobile phase viscosities in the range of 1.2–8.7 mPa⋅s. Mobile phases were viscosified through addition of sucrose, glycerol, or dextran. Change in mass transfer resistance, measured via van Deemter curves, was related to the change in diffusivity through viscosity. The analysis of different viscosifying agents emphasized the influence of viscosity inside the pores, rather than viscosity of the bulk phase. The viscosity inside the pores was calculated via the partition coefficient of each viscosifying agent. Based on the slopes of van Deemter curves, column dimensions were calculated for different scenarios, assuming a non-compressible stationary phase. Column volume remained constant with stream dilution from 8.7 mPa⋅s down to about 2.5 mPa⋅s. However, at the same time column geometry changed to thinner and longer columns with decreasing viscosity, in order to accommodate throughput and pressure drop. When diluting to even lower viscosities, column volume increased, since stream viscosity is less sensitive to stream concentration at the low viscosity range. These results are relevant to a wide range of industries utilizing weak interaction chromatography, especially those where the main driver of process development is cost reduction and where a trade-off between purity, yield, and costs has to be made.

Pepsin is the first protease that food proteins encounter in the digestive tract. However, most of the previous studies on the enzymatic kinetics of pepsin were based on the hydrolysis of small synthetic peptides, due to the limitations in methodology and the complexity of protein substrate. To better understand the role of pepsin in protein digestion, we used isothermal titration calorimetry to study the enzymatic kinetics of pepsin with bovine serum albumin as the substrate. We found that pepsin has a higher catalytic rate at lower pH, while its affinity to substrate is lower. At the same pH, pepsin has lower activity and affinity at higher ionic strengths. We found contrasting kinetic parameters for pepsin-catalyzed hydrolysis of bovine serum albumin and of small synthetic peptides. Time-dependent kinetics also showed that pepsin has lower efficiency towards intermediate peptides during hydrolysis.

Dry milling and subsequent sieving were evaluated as an alternative to the conventional wet extraction of quinoa (Chenopodium quinoa Willd.) proteins and starch. Specifically, quinoa sweet varieties have the potential to be dry fractionated. Dry fractionation of quinoa is an alternative and more sustainable route for producing protein-enriched and starch fractions than conventionally wet fractionation. Quinoa seeds were subjected to coarse grinding and subsequently sieved using different sizes of sieves to obtain fractions enriched in protein and starch. The protein-enriched fractions contained ∼32% proteins (32 g/100 g dry solids) while the starch-rich fractions contained 86–89% starch (86–89 g/100 g dry solids). The quinoa fractions were characterised and compared to wet-isolated starch and protein. The gelatinization temperature of the starch-rich fraction was influenced by the residual presence of proteins. The starch-rich fraction also had different pasting properties than starch isolate. The unheated protein-enriched fractions showed high water retention capacity and solubility, which could be potentially interesting to apply in gluten-free products.

The in vitro gastric digestibility of the quinoa variety Riobamba was investigated, especially the influence of the quinoa matrix. Dry-fractionated quinoa protein concentrate, which is just milled and sieved, was much better digestible than the same concentrate that was reconstituted from wet fractionated quinoa protein isolate, quinoa starch isolate, and quinoa fibre isolate. In the reconstituted concentrate, the presence of quinoa starch and fibre next to quinoa protein reduces its in vitro gastric digestibility significantly. However, the effect of starch is partially counteracted if the fibre is also present. While the effects of starch and fibre separately can be understood from the decrease of the accessibility for pepsin to hydrolyse proteins, due to the hydrated starch and fibre, we suspect that the synergistic effect of starch and fibre may be due to a relative reduction of the hydration of starch due to the presence of the also strongly hydrating fibre. We concluded that the presence of starch and fibre decreases the protein in vitro gastric digestibility. Therefore, the presence of fibre partially countered the decreased of the protein digestibility of starch. Heating of the matrices to 120 °C generally resulted in much lower digestion rates, due to extensive aggregation of the protein.

Quinoa protein was isolated from quinoa seeds using wet fractionation that resulted in a protein isolate (QPI) with a high protein purity of 87.1% (w/dw) and a protein yield of around 54%, and a dry fractionation method delivered a quinoa protein concentrate (QPC) with a purity of 27.8% (w/dw) and yield of around 47%. The dry fractionation process only involves milling and sieving and keeps the protein in its natural, native state. The aim was to study the in vitro gastric digestibility of both protein. Attention was paid to thermal pre-treatment of QPI and QPC. QPC showed significantly higher (p < .05) digestibility than QPI samples. The results were interpreted with a simple double exponential model. The fraction of easily digested protein in QPC is higher than for QPI. The better digestibility of the QPC was explained by the prevention of the formation of large aggregates during pre-heating of the protein.

The gastric digestion of proteins is influenced by the pH and the gastric pH fluctuates after food consumption. However, the dynamics of gastric pH still need to be quantitatively understood. Proteins in food strongly influences the gastric pH. Therefore, we studied the interaction between acid and proteins, including the buffer reaction and the acid diffusion in protein gels. The buffer capacity of proteins stems from its content of ionizable amino acid side groups. Based on this, we set up a model and method to parameterize the buffer capacity of proteins. Moreover, the liberated carboxyl and amino groups during enzymatic hydrolysis of protein can also contribute to the buffer capacity. While we expected protons to diffuse faster than pepsin, we found that the penetration distance of acid is comparable to that of pepsin. The buffer reaction caused the acid to concentrate tenfold in the gel compared to the bulk acid concentration. Therefore, we postulated that the buffer reaction reduces acid diffusivity in gels.

Protein is essential to human health, but its digestion kinetics in varied structures are not yet well understood. We previously found different kinetics of protein hydrolysis in solution and in gels, and we hypothesized that the difference stemmed from the steric hindrance of gel structure to the diffusion of pepsin and its hydrolysates. To better understand the pepsin diffusivity in food matrices and its effect on digestion, we determined the diffusivity of pepsin in water and in whey protein isolate (WPI) gels by fluorescence correlation spectroscopy (FCS). We estimated the pepsin concentration gradient during digestion based on the determined diffusivity, which showed that the pepsin is constrained within a thin layer from the gel surface. Gel composition analysis confirmed this constraint: peptides as protein fragments were observed only in the first 2 mm of the WPI gels after 6 h of in vitro gastric digestion. Scanning electron microscopy indicated that pepsin loosened the microstructure of whey protein gel surfaces, which may accelerate pepsin diffusion and consequently gel surface disintegration. We conclude that the mode of whey protein gel digestion is determined by the summed effect of diffusion limitation, hydrolysis rate and microstructure transformation.

This chapter reviews the mechanics of human digestion from an engineering viewpoint. First, digestive processes, including mastication, bolus transport and nutrient absorption, are paralleled to unit operations, such as milling, peristaltic pumping and membrane filtration. This allows an engineering analysis of the gut, which includes dimensional analysis and identification of the key parameters and phenomena that determine the rate and extent of food digestion. Then, the use of mathematical models and computational fluid dynamics are discussed for the study of digestive processes, in particular gastric and intestinal flow and mixing.

Process conditions that are applied to make structured soy-protein-based food commonly include high temperatures. Those conditions can induce protein oxidation, leading to a decrease in their susceptibility to proteolysis by digestive enzymes. We aimed to investigate the effects of thermomechanical processing on oxidation and in vitro gastric digestion of commercial soy protein ingredients. Samples were sheared at 100 to 140 °C and characterized for acid uptake, carbonyl content, electrophoresis, and surface hydrophobicity. The enzymatic hydrolysis was determined in simulated gastric conditions. Protein ingredients were already oxidized and showed higher surface hydrophobicity and hydrolysis rate compared with those of the processed matrices. However, no clear correlation between the level of carbonyls and the hydrolysis rate was found. Therefore, we conclude that gastric digestion is mostly driven by the matrix structure and composition and the available contact area between the substrate and proteolytic enzymes.

Fructooligosaccharides (FOS) were fractionated using nanofiltration cascades. Instead of creating one product and a residual stream, we report on configurations that create 3 separate product streams rich in: (1) monosaccharides (DP1), (2) DP3 and (3) DP ≥ 5. We developed a modular system allowing different operating pressures and membrane types at each stage. Two possible alternative configurations were assessed for a 3-stage cascade both experimentally and via simulation. The simulation was performed using a steady state model and was in a good agreement with the experimental data. Using the simulation model, the system was optimized towards 4 and 5 stage cascades. All designs were evaluated based on the purities and yields of 3 components of interest in the corresponding product streams. Selecting the correct set up, the cascade was able to reach maximum purity of monosaccharides to 66 wt% (from 9 wt%), DP3 to 33 wt% (from 24 wt%) and DP ≥ 5 to 54 wt% (from 34 wt%). Increasing the number of stages improved the maximum purities of the 3 fractions. However, a fifth stage did not increase the purification and the best purities were found using 4-stage rather than 5-stage cascades.

Open nanofiltration of mixtures of fructo-oligosaccharides was assessed by experiment and by modelling the overall permeation behaviour of 3 different membranes. The temperature effect was modelled using the steric pore model, incorporating the molecular volumetric expansion of fructo-oligosaccharides as solutes, the decrease in the solution viscosity and the volumetric expansion of the membrane with increasing temperature. The thermal expansion of the solute was described as a linear increase in the bare molecular volume plus a non-linear decrease in its hydration number. The viscosity reduction was modelled by incorporating the temperature as a variable into an existing exponential relation derived by Chirife and Buera. The thermal expansion of membranes was described with a linear increase in the pore size and a linear decrease in its hydrodynamic resistance. Although the purity of the oligosaccharide product was hardly affected by the temperature, the yield was much lower at higher temperatures. The yield can therefore be improved by decreasing the temperature while maintaining the product purity. This behaviour was also observed in a 3-stage filtration cascade. The temperature effect is closely related to the increase in fluxes with temperature, leading to a different split of the feed into permeate and retentate. In a membrane cascade, the lower yield with higher temperatures was seen most strongly at the top stage, and much less at the middle and lower stages, which can be explained by the configuration of the cascade.

Membrane cascades can fractionate fructooligosaccharides into 3 different fractions with varying degrees of polymerization (DP). In contrast to the traditional membrane system, membrane cascades have flexibility in configuration and setup for each stage. Apart from the improvement flexibility of the cascades provides, it raises problems related to multiple performance indicators and multiple process parameters. Therefore, new design criteria are required. We have designed an optimization approach for this multi-criteria problem. Eight configurations of cascaded membranes were built, measured and simulated to develop a design strategy. The performance of the separation process was evaluated by 10 different indicators: purities and yields for 3 different fractions and 4 separation factors between molecules with an adjacent DP. We found that the proposed configurations exceeded the performance of the previously reported 3-stage membrane cascade. Within those configurations, the cascade designs were able to increase the purity of (1) monosaccharides to 47% from 9%, (2) DP3 to 34% from 24% and (3) DP ≥ 5 to 77% from 34%. We also report a procedure to select a single optimum combination that compromises all performance indicators. This procedure systematically calculated the weights, which were then used to rank all feasible combinations and select the best one. In addition, a backward analysis using sensitivity coefficients was performed to pinpoint critical process parameters. Knowing these parameters, more targeted and more efficient improvements could be made. This approach is applicable for most integrated systems with multi-process variables and multi-performance indicators combining process modelling and multi-criteria decision making.

During gastric digestion, hydrolysis of proteins by pepsin contributes largely to the breakdown of protein-rich food. We hypothesized that the effect of pepsin is limited by its diffusivity, which is co-determined by the food structure and the local pH in the food during digestion. To investigate the principle mechanism of enzyme diffusion in food matrices, we used enhanced green fluorescent protein (EGFP) as probe to study the diffusivity of proteins in whey protein isolate gels, using fluorescence correlation spectroscopy (FCS). Gels made with different ionic strength showed distinctive elastic moduli but did not show differences in diffusivity of EGFP. Some models for diffusion in hydrogels yield good description of the obtained data, and can approximate the enzyme diffusion in diverse food matrices. However, the enzyme pepsin is more complicated than the probe EGFP, to yield more accurate predictions, electrostatic and enzyme-substrate interaction also need to be considered.

BACKGROUND: Membrane filtration has the potential for purification of oligosaccharide mixtures. However, single stage performance is limited by the relatively low yield and purity that can be achieved. This paper explores the potential of a three-stage pilot-scale spiral wound membrane filtration cascade for the removal of monosaccharides and disaccharides from a galacto-oligosaccharide mixture. RESULTS: Two three-stage cascaded configurations with GH membranes were tested, one ideal-like configuration (cascade 1) and one adapted configuration (cascade 2). Cascade 1 performed well with regard to oligosaccharide yield (e.g. yield 84%, purity 0.46), while cascade 2 performed better in terms of oligosaccharide purity (e.g. yield 62%, purity 0.58). This might be compared with conventional single stage concentration which resulted in an oligosaccharide yield of 56% with a purity of 0.53. In addition these cascade configurations were explored using different membranes (GE and GK membranes) in each stage. CONCLUSION: The yield and purity in the cascaded systems were higher than the yield and purity of a single stage system with volume reduction. The yield and purity can be further enhanced by using an inhomogeneous cascade with different membranes in each stage.

The adsorption of bovine serum albumin (BSA) to a chromatography resin with immobilised llama antibody fragments as affinity ligands was investigated. The maximum adsorption capacity of the affinity resin was 21.6 mg mL−1 with a Langmuir equilibrium constant of 20.4 mg mg−1. Using packed bed chromatography, BSA was adsorbed from pure BSA solutions. Recovery was achieved by desorption at pH 3. In experiments with initial BSA concentrations of 1 mg mL−1 and 0.1 mg mL−1, BSA could be concentrated to 6.9 and 7.7 mg mL−1, respectively. BSA was also isolated from filtered bovine cheese whey containing less than 0.1 mg mL−1 BSA. The purified BSA was in this case concentrated to 7.4 mg mL−1 BSA. The presence of other components in the feedstock did not alter the adsorption capacity of the affinity resin. The results show that high recovery combined with high purity can be obtained using affinity chromatography.

The objective of this study was to analyse the impact of the gel structure obtained by different heat-induced temperatures on the in vitro gastric digestibility at pH 2. To achieve this, gels were prepared from soy protein, pea protein, albumin from chicken egg white and whey protein isolate at varying temperatures (90, 120 and 140 °C) for 30 min. Gels were characterised prior to digestion via microstructure and SDS-PAGE analysis. Subsequently, the gastric digestion process was followed via the protein hydrolysis and HPSEC analysis up to 180 min. Peptides of different sizes (

In industrial liquid separation processes chromatography often has a key function in the optimization of yield and purity. For the design of an industrial system, chromatographic processes are generally simulated using mathematical models, tested and optimized at laboratory level, and then scaled up to pilot and subsequently industrial scale. To describe the system, experimental data and model data need to be fitted and extra column contribution must be determined. This paper describes the influence of extra-column volume on overall separation efficiency for lab scale and its impact on the design of large scale systems. Measurement of extra-column contribution was investigated in terms of mean retention time and variance using two different methods the commonly used zero dead volume connector and as an alternative the zero length column. Further a technique is presented to estimate extra-column contribution to band broadening for different injection volumes, velocities, and tracers based on representative measurements. When scaling up, often contribution of extra-column volume from laboratory equipment is neglected assuming to be on the safe side, however column efficiency is often lower than efficiency measured for the entire chromatographic system. Relation between system efficiency and column efficiency was investigated using laboratory data and the lumped kinetic model. Depending on the ratio of extra-column volume to retention volume in the system, deduced column efficiency was up to 20% smaller than overall system efficiency. This ratio revealed the misleading nature of the term efficiency loss, when describing influence of extra-column volume on column efficiency. A scheme, which relates the relative variance of the system to the relative extra-column volume, provided an assessment of under- or overestimation of column efficiency. In this article it is shown how scaling up a system based on laboratory data, where extra-column volume contribution is not accounted for, may severely overestimate column efficiency. This overestimation results in underestimated column dimensions at pilot and industrial scale, and hence underperformance of the industrial system. © 2017 Elsevier B.V.

To assess the potential use of ideal nanofiltration cascades for the industrial fractionation of oligosaccharides, simulations of single, three and five stage NF cascades were carried out. Three and five stage ideal cascades show significant improvement in separation with diafiltration compared to single stage systems. The calculations do imply different membrane areas in each stage of the cascade. The ratio of the sieving coefficients of a binary mixture over the membrane plays an important role in determining the relation between yield and purity in a cascade system. At high sieving coefficient ratios, both yield and purity increase concurrently in a three stage system, whereas at a low ratio of the sieving coefficients, the yield and purity become inversely proportional on the retentate side. In a five stage systems, both yield and purity become inversely proportional at high and low sieving coefficient ratios. A five stage cascade system installation would be optimal for most applications since at very low local separation factors sufficient separation and yield could be achieved.

Enzymatic wheat gluten hydrolysis at high solid concentrations is advantageous from an environmental and economic point of view. However, increased wheat gluten concentrations result in a concentration effect with a decreased hydrolysis rate at constant enzyme-to-substrate ratios and a decreased maximum attainable degree of hydrolysis (DH%). We here identified the underlying factors causing the concentration effect. Wheat gluten was hydrolyzed at solid concentrations from 4.4% to 70%. The decreased hydrolysis rate was present at all solid concentrations and at any time of the reaction. Mass transfer limitations, enzyme inhibition and water activity were shown to not cause this hydrolysis rate limitation up to 50% solids. However, the hydrolysis rate limitation can be, at least partly, explained by a second-order enzyme inactivation process. Furthermore, mass transfer impeded the hydrolysis above 60% solids. Addition of enzyme after 24 h at high solid concentrations scarcely increased the DH%, suggesting that the maximum attainable DH% decreases at high solid concentrations. Reduced enzyme activities caused by low water activities can explain this DH% limitation. Finally, a possible influence of the plastein reaction on the DH% limitation is discussed.

We report here that the usage of immobilized enzyme in a continuous packed bed reactor (PBR) can be a good alternative for GOS production instead of the traditional use of free enzyme in a batch reactor. The carbohydrate composition of the product of the PBR with immobilized enzyme was comparable to that of the batch reactor with free enzyme. The stability of the immobilized enzyme at a lactose concentration of 38% (w/v) and at 50 °C was very high: the half-life time of the immobilized enzyme was approximately 90 days. The enzymatic productivity of GOS production using immobilized enzyme in a PBR can be more than six times higher than that of GOS production with free enzyme in a batch reactor. Besides, when aiming for an equal volumetric productivity to the batch process in designing a PBR, the volume of the PBR can be much smaller than that of the batch reactor, depending on the enzyme dosage and the run time of a single batch.

The adsorption of bovine serum albumin (BSA) to an immobilized camelid-derived antibody fragment was investigated using six different activated resins, of which two are prototypes. The resins differed in base material, coupling chemistry and particle size. The adsorption, washing and desorption stage of the affinity chromatography process were taken into account. Dynamic binding capacities at 10% breakthrough ranged between 0.76 and 4.8 mg BSA/mL resin. The washing volume ranged between 2.9 and 10 column volumes. One of the resins did not concentrate BSA, while the highest concentration was 13-fold. We present a method to rank and weigh the properties of the resins to find the optimal resin to meet specific requirements. For three of the resins the adsorption flow rate was varied, while the washing and desorption flow rate was kept the same. The dynamic binding capacity decreased with increasing flow rate, as expected. For one resin, the washing volume remained constant, but for the others it decreased with increasing adsorption flow rate. The number of column volumes required to purify a given amount of BSA increases with increasing flow rate, which indicates that higher flow rates do not necessarily speed up the process.

The aim of this study was to evaluate the potential of NF membrane cascades for continuous oligosaccharide purification. Three different nanofiltration membranes were evaluated, and the best combination in terms of membrane type and process parameters was determined for two commercial oligosaccharide mixtures of fructooligosaccharides (FOS) and galactooligosaccharides (GOS). To represent the cascade mathematically, a dynamic model was built based on film theory and on measurements performed in single-stage conditions. The model predictability was demonstrated with experiments in a membrane cascade set-up. Considering an initial purity of 84% for FOS and 40.4% for GOS, the model predicted a maximum attainable purity of 94.9% and 46.7% for FOS and GOS respectively. A minimum yield of 90% was used as constraint during the optimisation process, in which the physical limitations of the set up were also taken into account. This paper demonstrates that the trade-off situation between purity and yield can be overcome by using cascade configuration, leading to an efficient separation that cannot be achieved by single-stage membrane systems.

The aim of this study was to determine the influence of heat processing on denaturation and digestibility properties of protein isolates obtained from sweet quinoa (Chenopodium quinoa Willd) at various extraction pH values (8, 9, 10 and 11). Pretreatment of suspensions of protein isolates at 60, 90 and 120 °C for 30 min led to protein denaturation and aggregation, which was enhanced at higher treatment temperatures. The in vitro gastric digestibility measured during 6 h was lower for protein extracts pre-treated at 90 and 120 °C compared to 60 °C. The digestibility decreased with increasing extraction pH, which could be ascribed to protein aggregation. Protein digestibility of the quinoa protein isolates was higher compared to wholemeal quinoa flour. We conclude that an interactive effect of processing temperature and extraction pH on in vitro gastric digestibility of quinoa protein isolates obtained at various extraction pH is observed. This gives a first indication of how the nutritional value of quinoa protein could be influenced by heat processing, protein extraction conditions and other grain components.

Processes such as chromatographic separation and nanofiltration can remove low molecular weight sugars from liquid mixtures of oligosaccharides. As an alternative for the separation of such liquid mixtures, we studied mass diffusion separation of such sugars in a microfluidic device with incorporated nanofiltration membranes. This separation method is based on differences between diffusivities of components and does not require high transmembrane pressures. The effects of channel depth and flow rate were studied in experiments. The key parameters selectivity and rejection increased with increasing channel depth due to increased external mass transfer limitations. Among the studied membranes, the obtained selectivities and rejections correlated to the specified retention values by the manufacturers. Compared to more conventional nanofiltration where high pressure forces solutes through membranes, we obtained corresponding selectivities and fluxes of only an order of magnitude smaller. Simulated results indicated that with optimized microchannel and membrane dimensions, the presented separation process can compete with currently available separation technologies.

Enzyme microreactors are used as research tools because of many advantages including low reagent consumption and the general notion that mass transfer restrictions are reduced. The use of microchannels can indeed shorten the characteristic mass transfer time but may also affect the productivity of the microreactor. To what extent mass transfer restrictions affect the reaction rate and the productivity is determined by parameters such as the enzyme properties, operating conditions, and dimensions of the microreactor. This article provides the correlations between these parameters for coflow enzyme microreactors obeying Michaelis–Menten kinetics. These correlations outline the design space based on reduced mass transfer restrictions and maximum productivity, respectively. The methodology that yields the design space provides a generic hands-on approach to optimally design coflow enzyme microreactors. We conclude that effectiveness is an important parameter for designing microreactors, but – depending on the reactor configuration – does not necessarily lead to optimized throughput. We demonstrate design windows to illustrate this effect and how to take advantage of it.

Horseradish peroxidase can induce the oxidative cross-linking of proteins through the radicalization of tyrosine residues and subsequent formation of dityrosine bonds. The dityrosine bond absorbs light at 318 nm which can be used to monitor in-line the peroxidase-catalyzed cross-linking of proteins in a microfluidic system. In this study calcium-depleted α-lactalbumin is used as model protein. To quantify the progress of the reaction, the absorbance increase at 318 nm was monitored in-line and compared with the amount of reacted monomeric α-lactalbumin as determined with size-exclusion chromatography (SEC) at various residence times. The increase in absorbance at 318 nm shows a logarithmic relation with the extent of reacted monomer. The logarithmic relation can be explained using a reaction model describing minimum and maximum formation of dityrosine cross-links to reacted monomer. Since the size distribution of reaction products was found to be reproducible, the absorbance increase at 318 nm can be used as a fast in-line screening method for the peroxidase-mediated cross-linking of proteins.

The stability of beta-glucosidase from the hyperthermophile Pyrococcus furiosus was studied as a function of pressure, temperature and pH. The conformational stability was monitored using FTIR spectroscopy, and the functional enzyme stability was monitored by inactivation studies. The enzyme proved to be highly piezostable and thermostable, with an unfolding pressure of 800 MPa at 85 degrees C. The tentative pressure-temperature stability diagram indicates that this enzyme is stabilized against thermal unfolding at low pressures. The activity measurements showed a two-step inactivation mechanism due to pressure that was most pronounced at lower temperatures. The first part of this inactivation took place at pressures below 300 MPa and was not visible as a conformational transition. The second transition in activity was concomitant with the conformational transition. An increase in pH from 5.5 to 6.5 was found to have a stabilizing effect.

A stochastic model was developed that was used to describe the formation and breakdown of all saccharides involved during alpha-amylolytic starch hydrolysis in time. This model is based on the subsite maps found in literature for Bacillus amyloliquefaciens alpha-amylase (BAA) and Bacillus licheniformis alpha-amylase (BLA). Carbohydrate substrates were modeled in a relatively simple two-dimensional matrix. The predicted weight fractions of carbohydrates ranging from glucose to heptasaccharides and the predicted dextrose equivalent showed the same trend and order of magnitude as the corresponding experimental values. However, the absolute values were not the same. In case a well-defined substrate such as maltohexaose was used, comparable differences between the experimental and simulated data were observed indicating that the substrate model for starch does not cause these deviations. After changing the subsite map of BLA and the ratio between the time required for a productive and a non-productive attack for BAA, a better agreement between the model data and the experimental data was observed. Although the model input should be improved for more accurate predictions, the model can already be used to gain knowledge about the concentrations of all carbohydrates during hydrolysis with an alpha-amylase. In addition, this model also seems to be applicable to other depolymerase-based systems.

The enzymatic hydrolysis of starch is usually carried out with 30–35 w/w% starch in water. Higher substrate concentrations (50–70 w/w%) were reached by using a twin-screw extruder for gelatinisation and for mixing enzyme with gelatinised starch prior to enzymatic hydrolysis in a batch reactor. The aim of this study was to determine which parameters are important for gelatinisation of wheat starch and to investigate the effects of different extrusion conditions on the enzymatic hydrolysis. After extrusion, the degree of gelatinisation was measured. During hydrolysis, the carbohydrate composition, the dextrose equivalent (DE) and the alpha-amylase activity were measured. Gelatinisation measurements showed that mechanical forces lowered the temperature required for complete gelatinisation. During hydrolysis experiments, high DEs were observed even if starch was not completely gelatinised during extrusion. Due to high substrate concentrations, the residual alpha-amylase activity remained high throughout enzymatic hydrolysis, although high temperatures were used. Increased substrate concentrations did not affect the carbohydrate composition of the product. Furthermore, the time required for the batch hydrolysis step could be varied by choosing a different enzyme-to-substrate ratio. This article provides a basis for detailed optimisation of this process to develop an industrial-scale process at high substrate concentrations.

The feasibility of using magnetic particles for in-line product isolation during enzymatic conversion was studied. A comparison was made between a process based on magnetic particles and a conventional adsorption column. The enzymatic reaction was described by two consecutive first-order reactions (synthesis and subsequent hydrolysis), while the adsorption of substrate and product was described by multicomponent Langmuir isotherms. The yield as well as synthesis/hydrolysis ratio were calculated for various system characteristics. The results show that magnetic particles are very effective when the affinity with the particles is specific and for enzymatic conversions involving low ratios of the rate of synthesis versus the rate of hydrolysis. For slow conversions and for low specific affinity molecules column separations are more appropriate.

A general procedure was developed to measure the degree of gelatinisation in samples over a broad concentration range. Measurements based on birefringence, DSC (Differential scanning calorimetry), X-ray and amylose–iodine complex formation were used. If a 10 w/w % wheat starch–water mixture was used, each method resulted in approximately the same degree of gelatinisation vs. temperature curve. In case the gelatinisation of a 60 w/w % wheat starch–water mixture was followed as a function of the temperature, each method resulted in a different degree of gelatinisation vs. temperature curve. DSC and X-ray measurements are preferred, because they can be used to determine when the final stage of the gelatinisation process has been completed. Birefringence and amylose–iodine complex formation measurements are suitable alternatives if DSC and X-ray equipment is not available, but will lead to different results. The differences between the methods can be explained by considering the phenomena that take place during the gelatinisation at limiting water conditions. Based on the experimental data obtained with DSC and X-ray measurements, the gelatinisation of 10 w/w % and 60 w/w % wheat starch–water mixtures started at the same temperature (approximately 50 °C). However, complete gelatinisation was reached at different temperatures (approximately 75 °C and 115 °C for, respectively, 10 w/w % and 60 w/w % wheat starch–water mixtures) according to the experimental DSC and X-ray data. These results are in accordance with independent DSC measurements that were carried out. The Flory equation was adapted to provide a quantitative explanation for the curves describing the degree of starch gelatinisation as a function of the starch–water ratio and the temperature. The gelatinisation curves that were obtained with the model are in good agreement with the experimentally determined curves. The parameters T m 0 , ΔHu and χ12 that resulted in the lowest sum of the squared residuals are 291 ± 63 °C, 29.2 ± 3.9 kJ/mol and 0.53 ± 0.05 (95% confidence interval). These values agree with other values reported in literature.

Adsorption processes using a packed bed reactor are often used for the purification of nutraceuticals. However, when dealing with untreated streams, this leads to fouling and consequent increase in pressure drop. This work compares the use of channeled monoliths, “honeycomb” structures, with a packed bed, both made of the same type of activated carbon. The intrinsic permeability and performance of both structures during the adsorption of a bioactive peptide from a crude hydrolyzate were studied. Breakthrough experiments were performed on both types of systems under similar conditions. The results showed similar productivity and dynamic adsorptive capacity for both structures at comparable linear velocity and residence time, but the packed bed showed a strong pressure drop increase during column saturation, revealing plugging of the column, especially at high velocities (short residence times). The channeled monoliths did not present any significant pressure drop increase, and were able to operate at high velocities which increase the range of application for these types of processes.