Search

Acid milk gels with fat (3.5 or 10%) or without fat containing 0.2–1% of whey proteins (WP) or 0.2–1.5% of protein aggregates (fractal aggregates, microgels and mixed casein/WP aggregates with an 80/20 ratio) were investigated. The fat-containing systems were homogenized and the systems were preheated (90 °C, 11 min) and acidified to a pH of 4.6 by adding glucono-delta-lactone. The protein composition of the fat droplet interface was characterized by SDS-PAGE and the textural and rheological properties, microstructure, and whey separation of acid networks were determined. WP and fractal aggregates showed the best ability to improve the textural properties and microstructure of all acid milk gels and reduce whey separation. Increasing the concentrations in microgels and mixed aggregates did not lead to an increase in gel firmness or generate a strong impact on the protein network, but a high concentration in mixed aggregates could reduce whey separation. The fat droplet interface was made almost exclusively of caseins, even if the fractal aggregates could also be adsorbed when the interfacial surface was increased either through fat content or through an increase in homogenization pressure. Adding proteins changes the textural properties of the acid gels in all the systems, mainly due to their role in the continuous phase.

International audience; Heat treatments induce changes in the protein structure in infant milk formulas (IMFs). The present study aims to investigate whether these structural modifications affect protein digestion. Model IMFs (1.3% proteins), with a bovine or a human whey protein profile, were unheated or heated at 67.5°C or 80°C to reach 65% of denaturation, resulting in six protein structures. IMFs were submitted to in vitro static gastro-intestinal digestion simulating infant conditions. During digestion, laser light scattering was performed to analyze IMF destabilization and SDS-PAGE, OPA assay and cation exchange chromatography were used to monitor proteolysis. Results showed that, during gastric digestion, α-lactalbumin and β-lactoglobulin were resistant to hydrolysis in a similar manner for all protein structures within IMFs (p > 0.05), while the heat-induced denaturation of lactoferrin significantly increased its susceptibility to hydrolysis. Casein hydrolysis was enhanced when the native casein micelle structure was modified, i.e. partially disintegrated in the presence of lactoferrin or covered by heat-denatured whey proteins. The IMF destabilization at the end of the gastric digestion varied with protein structures, with larger particle size for IMF containing native casein micelles. During intestinal digestion, the kinetics of protein hydrolysis varied with the IMF protein structures, particularly for IMFs containing denatured lactoferrin, exhibiting higher proteolysis degree (67.5°C and 80°C vs. unheated) and essential amino acid bioaccessibility (67.5°C vs. unheated). Overall, the protein structures, generated by modulating the whey protein profile and the heating conditions , impacted the IMF destabilization during the gastric phase and the proteolysis during the entire simulated infant digestion.

International audience; During filtration of skimmed milk, the casein micelles accumulate at the membrane surface and their presence has a crucial impact on the performances of the operation and on the products properties. Information is still scarce, especially at ‘low’ temperature (≈8-12°C), about how the casein micelles behave at the membrane surface and interact in such concentrated regimes. The aim of this work was to understand the structural organization and behavior of concentrated casein micelles accumulated at the membrane during crossflow filtration (concentration, sol-gel transition, cohesiveness and reversibility) with a focus on impact of temperature. The strategy consisted in combining in-situ measurements by Small-Angle X-Ray Scattering coupled to crossflow filtration, with ex-situ characterization of concentrated dispersions of casein micelles using osmotic stress technique and rheology. Swelling-re-dispersion experiments were carried out to characterize the cohesiveness of micelles concentrates. During in-situ filtration performed at constant pressure, increasing temperature (12, 25, 42°C) increased permeate flux (which was mainly attributed to the decrease of viscosity of permeate) and led to higher accumulation of casein micelles and a thicker deposit. The relaxation of pressure showed that casein micelles deposit was more cohesive at low temperature. With osmotic stress and rheology experiments, it was observed that compressibility (concentration reached at given osmotic pressure) and sol-gel transition concentration of casein micelles increased with the temperature increasing from 7 to 20°C. Moreover, swelling of casein micelles gels at 7°C resulted in lower redispersion rate as compared to the swelling at 20°C regardless of the gel formation temperature. It was also observed that casein micelles gels obtained at 7°C are more cohesive than those prepared at 20°C. This work shows that carrying out filtration at low temperature led to a lower accumulation of casein micelle (lower compressibility of casein micelles), but to a higher cohesiveness of the accumulated micelles (lower sol/gel transition and swelling properties). These results were attributed to physico-chemical modifications of casein micelles (voluminosity, hydration) under temperature changes.

International audience; Background: Today's consumers demand natural and healthy functional food products with clean labels and less added ingredients and exogenous hydrocolloids. The goal is to produce dairy foods only made by using milk ingredients, while maintaining good sensory properties. Thanks to their multifunctional properties, dairy proteins are used to enhanced the texture of dairy products. However, their addition into dairy products can also have a great influence on taste and aroma perception. Scope and approach: This review concentrates on recent information regarding the impact of dairy proteins in dairy products with a focus on yogurts. The aim of the review is to highlight recent findings relating the composition and the physico-chemical properties of yogurts to their sensory properties. Key findings and conclusions: The addition of dairy proteins in yogurts induces changes in their microstructure, which can explain the modifications of their texture properties. Modifying the protein content also tends to decrease the release of most flavor compounds, which can impact the overall aroma balance and change aroma perception. Modifications of texture and aroma perceptions can also be due to cognitive interactions. Dairy protein aggregates are particularly promising as dairy texture enhancers, however almost no sensory studies have been performed to investigate their impact on texture and aroma perception. More studies focusing on the impact of dairy protein aggregates on the organoleptic qualities of yogurts and dairy products are needed to obtain natural and low-fat products while maintaining a desirable texture quality.

This study aimed to perform Temporal Dominance of Sensations by modality (M-TDS) combined with a multi-intake approach to investigate texture and flavor perception in semi-solid products. Trained panelists (n = 15) evaluated fat-free strawberry yogurts enriched with functional proteins involving texture modifications. As yogurt is a semi-solid product, its in-mouth residence time is short. A multi-intake approach was therefore expected to give more reliable information about the sensory properties perceived by panelists. The two modalities of texture and flavor were analyzed separately to characterize the effect of added proteins. Trials were made according to an experimental design with two factors (protein type and concentration) and three levels each. Different statistical treatments, taking or not the temporality of attributes into account, were performed on standardized and non-standardized data. The implementation of M-TDS was essential to highlight differences of flavor perception in addition to the more evident texture modifications. The study of sensory trajectories evidenced that texture modifications, induced by the use of different whey proteins, slightly modified the perception of flavor and sweetness. The global flavor perception of the samples varied with the number of spoons, which particularly impacted the taste attributes. This study highlighted the importance of using M-TDS when studying texture and flavor in semi-solid products, and the relevance of the multi-intake approach to characterize flavor perception. This methodology enabled panelists to evidence both marked texture differences and subtler flavor modifications, and these useful data were emphasized by combining different statistical treatments.

International audience; In the industrialized world, dairy products constitute an important part of the diet, especially in northern Europe and North America. In these regions, milk products contribute around 30% of the total dietary protein supply and represent about 65% of the intake of animal protein. The protein content of cow’s milk ranges from 32 to 35 g/L. There are two major types of milk protein: the caseins (80%), which are represented by four distinct proteins (αs1-,αs2-, β-, and κ-caseins), and the whey proteins (20%), which are represented by proteins such as β-lactoglobulin, α-lactalbumin, and lactoferrin. These two families of proteins are opposite in terms of structure. Caseins exhibit a loose and highly flexible structure and are associated into a supramolecular structure called the micelle, whereas whey proteins have a globular, well-defined three-dimensional structure. These structural differences between the two families markedly affect the behavior of these proteins in the gastrointestinal tract and particularly their susceptibility to hydrolysis by the digestive enzymes. The nutritive value of proteins, including milk proteins, is generally associated with their capacity to provide two components: nitrogen (related to protein quantity) and essential amino acids (related to protein quality). The overall nutritional efficiency of protein is most commonly measured via nitrogen retention, which assesses protein retention. In terms of protein quality, the nutritive value is related to the amino acid composition and the bioavailability of these amino acids. The content and the bioavailability of indispensable amino acids, that is, those that cannot be synthesized in the body and consequently must be supplied through the diet, are of particular concern.

Milk contains two types of proteins: Caseins and whey proteins. There are four caseins: α/β/γ/κ caseins that formed micelles in milk. Whey proteins are globular proteins between 12 and 200 kDa: α-lactalbumin; β-lactoglobulin; Serum albumin; Immunoglobulin and Lactoferrin. Industrials fractionate milk to produce whey protein and casein isolate to make ingredients for the formulation of dairy products. Whey protein isolate can be heated to functionalize or pasteurize them. The objective of these three studies are to characterize the isolates of whey proteins and the impact of heating treatment on whey protein fractions and formulations using different FFF techniques: Asymmetrical-FFF; 2D AsFlFFF-RPLC and Centrifugal-FFF Whey protein isolate heated at high temperature were studied by Centrifugal-FFF coupled with MALS and RI. Aggregates with the lowest density and size elute in the void peak and represent 51% of the mass fraction. For the bigger one, there are two populations: one smaller than 1 micrometer and one higher, up to 4.5 micrometers. For size determination, the biggest population fractionated has been collected and analyzed by DLS in batch. Infant milk formulation without fat was investigated by Asymmetrical Flow FFF coupled with MALS; QELS and RI. The formulation contains bovine milk and isolate whey protein fractions. The technique separates the non-aggregated whey protein to the casein micelles. The addition of lactoferrin partially disintegrated the casein micelles. The formulations are heated at 67.5°C or 80°C. Native whey protein content decreases with the heat treatment, leading to their aggregation on with the casein micelles. The phenomena depend on the temperature and protein composition. At 80°C, shape ratio Rg/Rh of casein micelles increase with the temperature when the formulation is enriched in lactoferrin. The Rg and Rh decrease when the formulation is enriched in α-lactalbumin and lactoferrin. Transmission Electronic Microscopy are correlated with these results. At 80°C, whey protein aggregates are fixed outside the micelle for the first formulation and they are fixed inside the micelles for the second formulation. Whey protein isolate contains few quantities of caseins. The second objective of this study is to identify the presence of trace of caseins in whey protein isolate using 2D AsFlFFF-RP-HPLC. This technique allows to separate proteins by their size and by their polarity. A whey protein standard mix is studied: α-lactalbumin, β-lactoglobulin A and B, BSA, and 3 caseins. A 2D map is obtained with all separated proteins.

Microfiltration, MF is largely used in dairy industry to separate casein micelles from whey proteins. The weak performances of polymeric membranes compared to ceramic membranes are a restriction for their implementation in industry. Adding a diafiltration step is promising to improve polymeric MF performances, however most of the works have focused on the hydraulic performances and few have investigated the repercussion of diafiltration on protein separation. This work aims at investigating the impact of both the nature of solvent and the diavolume ratio on the MF efficiency. Pilot-scale MF using polymeric membrane (Synder, 800 kDa) was carried out on thermized skimmed milk at 12 °C with a volume reduction ratio, VRR from 2.5 to 3.2 and a transmembrane pressure of 0.7 bar. Continuous diafiltration was performed using reverse osmosis water (ROW) or milk ultrafiltrate (PUF). Permeation flux and β-lactoglobulin transmission were evaluated. During PUF-diafiltration of VRR-2.8-retentate, permeation flux is stable and β- lactoglobulin transmission declines by 26 % after 4.8 diavolumes. During ROW-diafiltration, permeation flux increases linearly until 2.9 diavolumes, then reaching an upper value. Simultaneously β-lactoglobulin transmission follows a parabolic curve with a maximum increase of 68 % for VRR-3.2-retentate and 10 % for VRR-2.5-retentate. According to Darcy’s law, the reduction of permeate viscosity during ROW-diafiltration leads to the improvement of permeation flux. This behavor is accentuated by the fouling layer relaxation consequent to the diminution of the ionic strength. In case of high repulsive forces in the concentration layer (low ionic strength), the transmission of negatively charged whey proteins is however restricted. The decrease of protein concentration during diafiltration may also limit whey protein transmission for high diavolume ratios. Diafiltration using ROW is recommended to increase transmission of whey proteins. However, volume of solvent has to be adapted to the VRR reached prior diafiltration to enable an optimum transmission

International audience; Hydrophobised whey protein particles were prepared through successive acetylation and heat treatment practices, and the particle characteristics were modulated by CaCl 2 supplementation. Then, the usefulness of the hydrophobised protein particles for emulsification of a docosahexaenoic acid-rich oil was compared with that of heat-denatured whey protein. Addition of CaCl 2 into hydrophobised whey protein resulted in smaller protein particles and lower z-potential and interfacial tension values. It also decreased the creaming stability of the consequent emulsions. It was argued that besides Ca 2þ-protein charge interactions, Cl e anions bind to the hydrophobised particles, and the Pickering stabilisation of oil does not rely on interfacial tension reduction. Compared with heat-denatured whey protein, hydrophobised whey protein afforded a lower protection to oil against oxidation; the peroxide value of the oil emulsified using hydrophobised protein was higher during storage.

Introduction et but de l’étude: La pepsine, sécrétée dans l’estomac, est la première protéase rencontrée par les protéines alimentaires au cours de la digestion. L’estomac acidifie lentement les bols alimentaires et il a été montré que le pH gastrique a un impact majeur sur les cinétiques de protéolyse 1.En effet, il est généralement admis que la pepsine n’est active qu’à pH acide, pour pH ≤ 5, avec un maximum d’activité autour de pH 2 pour beaucoup de substrats 2.Elle est également représentée comme un diacide 3, ce qui semble vrai lorsqu’elle agit sur l’hémoglobine, substrat de référence pour mesurer son activité. Ce comportement en fonction du pH n’a pas été validé sur des mélanges de protéines alimentaires telles que les caséines, principales protéines du lait, ou les protéines du blanc d’œuf. L’objectif principal de cette étude est de comparer les profils d’activité de la pepsine en fonction du pH sur les caséines du lait de vache (CA) et les protéines du blanc d’œuf (BO).Matériel et méthodes: Les cinétiques d’hydrolyse de protéines de BO entier et de micelles de CA par la pepsine, à différents pH compris entre 1 et 8, ont été suivies au pH-stat pendant 2h, avec quantification du degré d’hydrolyse final (DH) par la méthode OPA. Les vitesses initiales ont été calculées à partir de ces cinétiques entre 0 et 5 minutes. Les concentrations en substrat et en pepsine étaient les mêmes pour chaque expérience, et l’environnement salin était celui du protocole INFOGEST 4. Les CA se présentent en structures supramoléculaires, les micelles, qui précipitent à pH acide. Une étape d’acidification à pH 4 des CA a permis d’étudier des suspensions d’agrégats tous formés dans les mêmes conditions. L’aspect et la taille de ces agrégats ont été observés par microscopie confocale et par morpho-granulométrie.Résultats et Analyse statistique: Les cinétiques d’hydrolyse obtenues suivent, dans leur majorité, des profils de type loi de puissance en fonction du temps. Pour les protéines de BO, le DH final est maximal à pH1 (14,4%) et diminue de manière linéaire en fonction du pH,jusqu’à atteindre des valeurs négligeables pour les pH ≥6. A l’inverse, les valeurs de DH finaux sur les CA ne sont pas significativement différentes sur la gamme de pH 1 à 5 (7,5 ± 1,4 %) puis diminuent à partir de pH 6 et atteignent des valeurs négligeables à pH 8.Le rapport entre la vitesse d’hydrolyse initiale et le DH final des CA est supérieur à celui des protéines du BO quel que soit le pH, suggérant une cinétique d’hydrolyse plus rapide pour les CA.Ces résultats montrent que les caséines du lait,dont le pH se situe un peu au-dessus de 6, n’ont pas besoin d’être longuement acidifiées pour que l’hydrolyse par la pepsine puisse commencer dans l’estomac, atteignant une activité quasi optimale de la pepsine dès pH 5.Conclusion: Les cinétiques d’hydrolyse par la pepsine en fonction du pH diffèrent donc entre un substrat de type protéines globulaires comme l’œuf, et les caséines du lait. Le profil d’activité de la pepsine en fonction du pH est également différent de celui du standard hémoglobine. En outre, contrairement à ce qui est généralement admis, la pepsine peut hydrolyser certains substrats à des pH plus hauts que pH 5.

National audience; Introduction et but de l’étude Le processus de fabrication des préparations pour nourrissons (PPNs) implique des traitements thermiques, modifiant les propriétés physico-chimiques des protéines et éventuellement leur cinétique de protéolyse. L'objectif de l'étude consiste en l’analyse des effets de la composition et des structures protéiques thermo-induites des PPNs sur la cinétique de protéolyse. Matériel et méthodes Deux PPNs modèles ont été formulées à 1,3% de protéines. Elles se distinguent par la qualité des protéines du lactosérum (α-lactalbumine, β-lactoglobuline et lactoferrine) : la PPN « standard » mime la composition protéique de la majorité des PPNs 1er âge commercialisées et la PPN « LF+ ALA+ » celle du lait maternel. Les PPNs ont été traitées thermiquement à 67,5°C ou 80°C avec un taux de dénaturation des protéines du lactosérum identique de 65%. Les PPNs sans traitement thermique ont servies de référence. Les structures protéiques thermo-induites ont été caractérisées par électrophorèse, fractionnement d’écoulement de champ couplé à la diffusion de lumière multi-angle et microscopie électronique. D’autre part, les PPNs ont été soumises à la digestion gastro-intestinale en condition statique in vitro au stade nouveau-né à terme (n = 3). Le degré d’hydrolyse a été mesuré par dosage des amines primaires, la cinétique de protéolyse suivie par électrophorèse et l’évolution de la structure au cours de la digestion par granulométrie à diffraction laser. Résultats et analyse statistique A taux de dénaturation protéique équivalent, les caractéristiques morphologiques des structures protéiques thermo-induites diffèrent selon la composition protéique des PPNs et la température de traitement thermique. La formation d’agrégats solubles a été observée uniquement pour la PPN « standard », tandis que la fixation de protéines dénaturées sur les micelles de caséines a été constatée pour les deux PPNs. Les agrégats formés à 67,5°C ont une plus petite masse pour la PPN « standard » et un plus petit diamètre pour la PPN « LF+ ALA+ » que ceux formés à 80°C. Au cours de la phase gastrique, le traitement thermique accélère la cinétique de protéolyse des caséines de la PPN « standard » contrairement à la PPN « LF+ ALA+ ». La cinétique de protéolyse de l’α-lactalbumine dépend de la composition protéique des PPNs et de la température de chauffage de la PPN « LF+ ALA+ ». La protéolyse de la lactoferrine est fortement affectée par le traitement thermique. Les entités protéiques formées par agrégation en fin de phase gastrique sont affectées par la composition protéique des PPNs et le traitement thermique. Conclusion Notre étude a permis de mettre en évidence les effets de la composition protéique des PPNs et des structures protéiques thermo-induites sur les cinétiques de protéolyse. Ces deux paramètres sont donc à prendre en compte pour l’optimisation des PPNs en termes de stratégie nutritionnelle.

National audience; La fortification du lait de femme (LF) est une pratique courante dans l’alimentation des prématurés, pour les aider à rattraper leur retard de croissance. Les fortifiants, généralement à base de protéines de lait de vache hydrolysées (FV), peuvent poser des problèmes d’intolérances alimentaires de par l’immaturité du système digestif du prématuré. Un nouveau fortifiant à base de protéines entières de lait d'ânesse (FA) a été mis au point et testé lors d’un essai clinique randomisé chez les prématurés. Il a été démontré que la fortification de LF par FA réduisait l'intolérance alimentaire, les résidus gastriques bilieux et les épisodes de vomissements par rapport à FV, contenant des protéines hydrolysées. Le but de ce travail était d’évaluer comment ces différents fortifiants (FA, FV) affectaient la digestion des macronutriments (protéines et lipides) du LF. Matériel et méthodes: Du LF d’enfant prématuré, enrichi avec FA ou FV en conditions isoprotéique et isocalorique, a été digéré à l’aide d’un modèle de digestion in vitro dynamique adapté aux conditions digestives du nouveau-né prématuré. Les échantillons ont été prélevés à différents temps de digestion gastro-intestinale et ont été caractérisés en termes de protéolyse par suivi en électrophorèse et par mesure du degré d’hydrolyse (technique OPA) et de lipolyse par chromatographie en couche mince. La microstructure a été suivi au cours de la digestion par granulométrie à diffraction laser (distribution de taille de particules) et microscopie confocale. Résultats et Analyse statistique : Les cinétiques d’hydrolyse des protéines majoritaires de LF ne différaient pas entre les laits au cours de la digestion. Concernant le degré d’hydrolyse global, il était plus élevé avant digestion pour LF+FV que pour LF+FA de par la présence de protéines hydrolysées dans FV. Cette différence était maintenue entre les deux laits au cours de la digestion, mais l’écart était réduit en fin de phase intestinale. Le degré de lipolyse augmentait de la même manière pour les deux types de fortifications au cours de la digestion. De gros agrégats étaient présent dans LF+FA avant et pendant la digestion gastrique. La taille de ces particules était réduite après 60 min de digestion gastrique. Dans la phase intestinale, l'évolution de la taille des particules était similaire entre LF+FA et LF+FV. Conclusion: La libération des peptides et acides aminés est ralentie pour le lait de femme fortifié avec les protéines entières de lait d’ânesse par rapport à celui fortifié aux protéines hydrolysées de lait bovin. Ceci sera compléter par le dosage des peptides et acides aminés libérés au cours de la digestion. Apporter des protéines entières au nouveau-né avec le FA pourrait d’apporter au prématuré les acides aminés de manière plus progressive, et pourrait ainsi avoir un intérêt physiologique bénéfique pour la maturation du système digestif et immunitaire

International audience; Understanding how proteins stabilize amorphous calcium orthophosphate (ACP) phases is of great importance in biology and for pharmaceutical or food applications. Until now, most of the former investigations about ACP−protein stability and equilibrium were performed under conditions where ACP colloidal nanoclusters are surrounded by low to moderate concentrations of peptides or proteins (15−30 g L−1). As a result, thequestion of ACP−protein interactions in highly concentrated protein systems has clearly been overlooked, whereas it corresponds to actual industrial conditions such as drying or membrane filtration in the dairy industry for instance. In this study, the structure of an ACP phase is monitored in association with one model phosphorylated protein (casein) using solid-state nuclear magnetic resonance (ssNMR) under two conditions of high protein concentration (300 and 400 g L−1). At both concentrations and at 25 °C, it is found that the caseins maintain the mineral phase in an amorphous form with no detectable influence on its structure or size. Interestingly, and in both cases, a significant amount of the nonphosphorylated side chains interacts with ACP through hydrogen bonds. The number of these interacting side chains is found to be higher at the highest casein concentration. At 45 °C, which is a destabilizing temperature of ACP under protein-free conditions, the amorphous structure of the mineral phase is partially transformed at a casein concentration of 300 g L−1, while it remains almost intact at a casein concentration of 400 g L−1. Therefore, these results clearly indicate that increasing the concentration of proteins favors ACP−protein interactions and stabilizes the ACP clusters more efficiently.

National audience

International audience; The production of milk powders is a strategic sector of the dairy industry. Due to its complex macromolecular composition, shedding light on the mechanisms governing milk drying constitutes a main industrial concern and a scientific challenge at the same time. In this study, we used the single droplet[1] and the directional drying[2] approaches to investigate the evaporation of colloidal mixes of the two major milk proteins (whey protein isolates and native phosphocaseinates). Notably, we observed the crack development in the samples to evaluate the competitive impact of the physico-chemical properties of whey proteins (rigid globular shape) and casein micelles (deformable sponge-like structure) on the drying process. Our outcomes are potentially useful to better control the functional properties of milk powders. Furthermore, they contribute to improve the understanding of the drying mechanisms in binary colloidal systems, which are still far from being fully elucidated.

International audience; Spray drying is a well-established process, but the phenomena ruling droplet-to-particle transition remain black box, resulting in variable powder properties and drying performances. Therefore, understanding precisely how the particle is formed and how to control its properties remain a major challenge. This work aims to breakdown the complexity of the drying phenomena using an exploratory multi-scale approach. Particle formation of different milk proteins (whey proteins and casein micelles) was investigated using different experimental systems (single pendant droplet, mono-dispersed droplets and spraying cone droplets) in controlled drying environments. The results showed that the drying of a single protein droplet included three distinct stages highlighting the occurrence of specific morphological events (constant rate shrinkage, buckling instability, vacuole nucleation). According to the type of proteins, these stages differed in drying kinetics and droplet dynamics, leading to characteristic and reproducible particle shapes irrespective of the droplet configuration and the drying conditions. These different drying behaviors were related to specific skin formation mechanisms at droplet air-liquid interface and consequent responses of the protein material to internal stresses. In this work, we aimed at understanding how the different physico-chemical properties of whey proteins (rigid globular shape) and casein micelles (deformable sponge-like structure) affect the mechanisms of formation of such a skin, and from this the particle final shape. However, the full understanding of the drying process of biocolloids is far to be achieved and, notably, the impact of solute properties on the morphological characteristics of the evaporating droplets, such as colloid segregation, skin formation and crack pattern development is not elucidated yet. To this aim, the use of model colloidal solutions, whose rheological behavior is more easily deducible, could represent a significant boost. Consequently, we compared the drying of droplets of whey proteins and casein micelles to that of dispersions of silica particles and polymer-coated silica particles, respectively. The mechanical behavior of such biological colloids and model silica dispersions was investigated through the analysis of crack formation, and the measurements of their mechanical properties using indentation testing. The study reveals numerous analogies between dairy and corresponding model systems, thus confirming these latter as powerful tools to highlight the signature of the matter on the drying process and at the molecular scale. The outcomes presented in this work represent a first step to shed light on the physics of drying dairy droplets; they open the road to further investigation about mixes of milk proteins. Moreover, the observation of mixtures of dairy colloids would contribute to improve the knowledge about the evaporation of colloidal polydisperse systems, which represents a current scientific challenge.

Les connaissances actuelles sur la microfiltration tangentielle 0.1 µm de lait écrémé (MF), un procédé de fractionnement des deux grands types de protéines du lait, sont principalement basées sur le savoir-faire des opérateurs et les connaissances expertes disponibles. L’optimisation multiobjectif du procédé requiert dans un premier temps de formuler le problème multiobjectif ; cette étape est centrale car c’est à travers elle que les objectifs sont explicités. Pour la MF, les données de la littérature sont insuffisantes à la représentation et à l’optimisation en simultané des objectifs contradictoires de la MF (maximiser la productivité et les propriétés des fractions obtenues et minimiser les coûts économiques et l'impact environnemental). Aucun modèle de simulation de la MF n’existe, en raison de la complexité du lait et des mécanismes de transformation mis en jeu lors de la filtration, et ce manque de modèle est un frein à l’optimisation multiobjectif du procédé [1]. La formulation du problème peut donc passer par un recueil de connaissances auprès d’experts, comme cela a été montré dans d’autres domaines [2]. L’objectif de ce travail est d’intégrer les connaissances expertes à la formulation du problème d’optimisation. Afin de faciliter le recueil de connaissances et l’identification des experts, le domaine de la MF a été divisé en neuf sous-domaines thématiques (Fig.1). Les connaissances et savoir-faire sont recueillis par la technique de l’interview semi-structurée [3] et sont représentés sous forme de cartes causales (Fig.2) [4]. Pour réaliser le recueil de connaissance, les objectifs de l’optimisation ont tout d’abord été définis. En prenant en compte les problématiques industrielles, les experts ont défini cinq objectifs -maximiser le rendement de récupération des protéines dans le perméat de la MF, maximiser la composition des fractions de rétentat et de perméat, minimiser les coûts d'investissement et de fonctionnement- auxquels il a été rajouté la minimisation de l’impact environnemental. Le problème d’optimisation de la MF est multidisciplinaire et les connaissances nécessaires à la modélisation et la résolution du problème sont partagées entre plusieurs experts. Pour chaque sous-domaine, une carte causale est établie par expert puis validée par ce dernier. Les cartes d’un même sous-domaine sont fusionnées, discutées et validées par le groupe d’experts concernés. La carte causale globale, obtenue de la même façon par fusion des cartes des sous-domaines, est le modèle de connaissances qui sera la base de la modélisation des objectifs de l’optimisation. Pour chaque sous-domaine, entre deux et cinq experts –industriels laitiers, équipementiers/assembliers et/ou chercheurs- ont été interviewés afin de recueillir les variables qui influencent les objectifs de l’optimisation. Neuf experts au total ont été interviewés avec en moyenne 90 minutes par expert. La carte causale globale comporte 152 variables et 318 relations d’influences. L’intégration des connaissances expertes sur les objectifs de l'optimisation et les variables qu'ils mettent en œuvre a permis d’identifier, via les cartes causales établies, les relations d’influence entre les variables et entre les variables et les objectifs. Elle a aussi permis d’identifier des trous de connaissances à combler par des actions de recherche, tel que l’impact de la diafiltration sur les caractéristiques des fractions obtenues, avant de pouvoir considérer l’opération de fractionnement dans son intégralité. La mise en équations du modèle de connaissance permettra d’optimiser en simultané les objectifs contradictoires de la MF afin de trouver les meilleurs compromis Pareto-optimaux dont certaines pourraient correspondre à des innovations de ruptures.

National audience; The Toulouse FFF Center (TFFFC) project aim is the development of a shared research platform for characterization of complex samples using Field-Flow Fractionation technologies. The project is built around collaborative projects with partners from various fields of activities such as agrifood, polymer chemistry, renewable carbon valorization and analytical equipment. The financial support of the Region Occitanie, European funds and industrial partners make possible the creation of an analytical platform dedicated to FFF technologies unique in Europe and to develop innovative approaches and tools allowing the characterization of complex samples by these Field-Flow Fractionation technologies. The performance of these techniques will be evaluated for the characterization various nano and micro-structures like proteins. We will also develop analytical methodologies using different FFF techniques to better characterize complex milk protein samples used as agro-food texturing agents and lignocellulosic materials, a product of interest in renewable carbon chemistry. These projects will be carried out in collaboration with industrial partners.

International audience; This paper aims at presenting a consistent Life Cycle assessment of the environmental impact of an industrialmilk protein fractionation process at the level of unit processes (gate to gate approach), focusing on the effectivecontribution of both the production and the cleaning steps. The precise assessment of the impact of cleaningstages could indeed constitute key issues for the optimisation and management of food plants. The results showthat the production phase accounts for approximately two-thirds of the environmental impact of the entire foodmanufacturing process, and the cleaning phase for more than 30% of the total impact. The latter is, however,underestimated due to methodological drawbacks and partly unavailable data. Among both the production andthe cleaning stages the most impacting unit processes are the membrane and the thermal operations mainly dueto the single-use of cleaning solutions and several temperature changes. Hence, for reduction of the overallenvironmental impact of the milk protein fractionation process it is advisable to primarily focus on improvementand optimisation of the microfiltration operation and secondly, on the thermal units.

International audience; The objectives of the study were to: 1 Determine whether gastric emptying of an isoproteic solution of CM and CS are different or not (exp. 1) 2 Characterize the structure of the resulting chyme and determine if CM and CS are differently metabolized (exp. 2)

Modelling and optimizing food processes is a complicated task due to the high complexity of the food product itself, the lack of knowledge concerning mechanisms limiting process performances [1], and to the heterogeneity of the involved variables (ordinal, cardinal, discrete or continuous variables). This is the case for skim milk crossflow microfiltration with 0.1 μm pore size (MF 0.1 μm). This operation is commonly used in dairy industry to separate proteins: native casein micelles (retentate) are used in cheese making while serum proteins (permeate) are mainly used in food formulations for specific populations (elderly people, infants, etc). Despite the high interest in MF in the dairy sector, this process is not optimized by considering the different conflicting objectives of the industry such as maximizing protein recovery in the permeate fraction while minimizing economic costs. The choices of membrane configurations, processing designs and operating conditions are based on the know-how of operators and the available expert knowledge. This knowledge is partial and does not take into account links between the process operating variables, product state variables and optimization objectives. When the optimization of MF 0.1 μm is performed, it is often described as mono-objective and empirical problem [2] or as study cases showing the influence of one variable on a group of chosen variables [3]. This article presents a methodology for collecting, collating and representing expert knowledge necessary to the design of food processing multi-objective optimization problems focusing on skim milk MF.2. MethodsThe proposed methodology has four steps: formulation of the multi-objective optimization problem, modelling optimization objectives as functions, optimization of objective functions and choice of a solution to implement [4]. In this work, the focus is on the problem formulation. TheMF 0.1 μm optimization problem is multidisciplinary and the knowledges necessary to formulate and solve the problem are shared between several experts. By dividing the problem domain into domains of knowledge, the identification of experts and the knowledge collection are easier [5]. The semi-structured interview was used to elicit know-how and knowledge from experts. The influence relations between variables and objectives were collected from the interviews and represented as causal maps. Graphical representation allows expert to exchange easily about influence relations. After validations, maps of the same knowledge domain were merged, discussed and validated by the group of the domain experts. Maps represent the knowledge models to use for formulating the optimization objectives as functions from the literature or experimental data.3. Results and discussionIn this work, the operational objective is to be able to compare optimized microfiltration systems within a global process that can accept different membranes, processing designs and operating conditions. Optimization objectives were defined with the stakeholders: ten experts and decision makers. First, ten objectives were defined and reduced to six regarding industrial issues (maximization of the retentate composition, permeate composition, permeate protein recovery and stability of the permeation flux and minimization of the investment cost and operating cost). The MF 0.1 μm optimization problem domain was divided into nine knowledge domains: membrane properties, performance of the filtration, operating variables, permeate fraction, retentate fraction, process design, costs and environmental impacts. Each knowledge domain benefited from two to seven experts. Knowledge was collected during 36 interviews representing 14h30 of audio recording. For each knowledge domain, a causal map by expert was established and the whole merged as the causal map of the domain. After validations, each domain causal maps were discussed and validated by the different domain experts (industrialists, scientists, equipment manufacturers). One very important point of this step was to define a shared glossary and a generic process scheme representative of all the specific schemes. By the end, the extracted causal relations between variables were represented in 9 maps using approximately 250 variables.4. ConclusionsThe optimization problem studied is complex with conflicting objectives and heterogeneous variables. The integration of expert knowledge allows the identification of relevant objectives regarding to industrial issues. The MF process is considered from conception to performances with the identification of lack of knowledge in order to plan new experiments. Causal maps allow an easy to understand representation of the influences between the variables and make it possible to establish which variables influence which objectives. Based on these maps, the objective functions will be formalized allowing to perform the optimization.

National audience

National audience; Ultra- and microfiltration of skim milk are widely used in the dairy industry for concentration and the fractionation of milk proteins. However their performances are limited by formation of a concentrated fouling layer of casein micelles at the membrane surface. Some precedent results on the possibly effect of operating conditions have been highlighted by Qu et al. (2012) and Jin et al. (2014). Nevertheless, there is a lack of knowledge on the behaviour of concentrated layer (sol-gel transition and reversibility) at different operating conditions. Consequently, the aim of this work is to elucidate the local structure and time dependant organization of casein micelles fouling layer under various operating conditions thanks to in-situ small-angle X-ray scattering SAXS during cross-flow ultrafiltration. This allowed to (i) distinguish the thickness and local solid concentration profiles in flowing (sol) and stagnant (gel) parts of the fouling layer; (ii) discriminate the effect of time, transmembrane pressure TMP, shear and temperature on the sol-gel transition; (iii) study the reversibility of the sol-gel transition under the moderate laminar cross flow. Material and Methods Dispersion with initial casein micelle concentration of 54.7 g.L-1 (2.1 C0 with C0 = 26 g.L-1) was prepared from casein micelle powder (Promilk 802 B, Ingredia, Arras, France) dispersed in preliminary obtained milk ultrafiltrate (Qu et al., 2012). The dispersions were prepared by thoroughly mixing the casein powder in a UF permeate for 15h at 35°C. SAXS experiments were performed at ID02 High Brilliance beamline of ESRF (Grenoble, France) with the help of specially developed SAXScross-flow ultrafiltration cell (Jin et al., 2014). Polyethersulfone ultrafiltration membrane with nominal MWCO of 100 kDa (Pleiade, Orelis Environnement, France) was used for filtration. Those experiments allowed to access to the structural organization at nanometer length scales and concentration profiles at the vicinity of the membrane surface. The profiles of casein micelles concentration in the fouling layer (local casein concentration versus distance to membrane) were obtained with spatial resolution of 20 μm. The impact of transmembrane pressure (10 and 110 kPa), cross-flow velocity (0.03, 0.05 and 0.15m.s-1) and, notably, temperature (12, 25 and 44°C) on the local structure of fouling layer was studied during different stages of filtration (layer formation → relaxation → erosion). Results and Discussion Filtration run in figure 1 was carried out at v = 0.03m.s-1 and TMP = 110 kPa. A second step of relaxation was performed by reducing the TMP at 10 kPa. Finally, erosion step at 10 kPa and 0.15m.s-1 was applied to try to eliminate the fouling layer. Such conditions allow to build a thick casein micelle layer of around 125 μm (Fig. 1) with a maximal concentration of 7.5 C0 (around 195 g.L-1) close to the membrane (z = 30 μm) above the sol- gel transition (180g/L), according to Bouchoux et al. (2009) and Jin et al. (2014). The SAXS analysis revealed that this highly concentrated layer is mostly reversible since it lasts a 50-70 μm of casein micelles layer reaching 3.5 C0 (around 91 g.L-1) close to the membrane. The effect of temperature on the concentration profile and its reversibility has been studied for three independent filtrations (Fig. 2). Concentrated layer thickness increased with temperature: from ∼100 μm at 12°C to 150 μm at 25°C and 200 μm at 44°C (after 183 ± 9min of filtration) and decreased with shear. However, the concentration distribution profile did not evolve with layer thickness. A maximal concentration is observed for filtration performed at 25°C. Irreversibility is similar for all the temperatures since a 3.5 C0 (around 91 g.L-1) fouling layer close to the membrane is lasting after pressure relaxation. Conclusion In-situ SAXS measurements during cross-flow ultrafiltration of casein micelles allowed to follow the formation and reversibility of concentration polarization layers as a function of time, TMP, cross-flow and temperature conditions. It leads to identify sol-gel transition of casein micelles near the membrane surface and to distinguish which part was reversible and eliminated by TMP relaxation. Those results are important for understanding the mechanisms of concentration during cross-flow filtration and they could be used to adjust the filtration parameters

Unravelling the mechanisms of food digestion to improve our knowledge of the effect of food on human health. 4. Simposio de Alimentos e Nutriçao

Les protéines solubles du lait pour texturer les aliments. journée scientifique "science des aliments et valorisation des bio-produits"

Mécanismes de formation de l’interface au cours du procédé de séchage : approche multiéchelles appliquée aux poudres laitières. journée scientifique "science des aliments et valorisation des bio-produits"

Fouling of falling film evaporators during the concentration of wheys . STLOpendays

Balance of inter and intra casein micelle interactions during enzymatic coagulation of milk . STLOpendays

Optimization of membrane processes for fractionation of dairy proteins. STLOpendays

understanding properties of casein micelles concentrated layers during crossflow filtration. STLOpendays

Understanding and modelling of dairy protein drying based on single and monodisperse droplet experiments . STLOpendays

International audience; The evaporation of colloidal solutions is frequently observed in nature and in everyday life. The investigation of the mechanisms taking place during the desiccation of biological fluids is currently a scientific challenge with potential biomedical and industrial applications. In the last few decades, seminal works have been performed mostly on dried droplets of saliva, urine and plasma. However, the full understanding of the drying process in biocolloids is far from being achieved and, notably, the impact of solute properties on the morphological characteristics of the evaporating droplets, such as colloid segregation, skin formation and crack pattern development, is still to be elucidated. For this purpose, the use of model colloidal solutions, whose rheological behavior is more easily deducible, could represent a significant boost. In this work, we compare the drying of droplets of whey proteins and casein micelles, the two main milk protein classes, to that of dispersions of silica particles and polymer-coated silica particles, respectively. The mechanical behavior of such biological colloids and model silica dispersions was investigated through the analysis of crack formation, and the measurements of their mechanical properties using indentation testing. The study reveals numerous analogies between dairy and the corresponding model systems, thus confirming the latter as a plausible powerful tool to highlight the signature of the matter at the molecular scale during the drying process

Classiquement, la texture des desserts lactés à pH neutre, est obtenue grâce à des épaississants ou des gélifiants. Dans certaines conditions, des agrégats de protéines laitières sont également capables de connecter les gouttelettes de l’émulsion et de texturer le système. Au cours de ces travaux, nous avons étudié la capacité de différents agrégats à texturer des émulsions tout en faisant le lien avec leurs propriétés d’adsorption et d’étalement à l’interface dans un mélange complexe contenant aussi des protéines solubles. L’objectif était de contrôler la compétition d’adsorption entre ces différentes espèces et au final, la composition interfaciale, afin de maîtriser la texture des émulsions. Les résultats montrent la possibilité de texturer les émulsions pendant l’homogénéisation (gel instantané) mais également au cours du stockage (gel différé), et ce, même pour des teneurs en matière grasse faible.

National audience

National audience; Propriétés interfaciales et moussantes des protéines. journée scientifique "science des aliments et valorisation des bio-produits"

Part of special issue: The 5th International Conference on Food Digestion - Edited by Didier Dupont, Pasquale Ferranti, Alan Mackie 5. International Conference on Food Digestion (INFOGEST) ; Rennes (France) - (2017-04-04 - 2017-04-06) / Conférence; In the frame of the COST action INFOGEST, a static in vitro digestion protocol has been elaborated aiming at the improvement of data comparability by harmonizing the experimental conditions. The success in harmonization was confirmed with inter-laboratory trials using skim milk powder as a standardized model food. Moreover, the physiological relevance of the gastric and intestinal endpoints of the static digestion protocol was demonstrated in a pig in vivo trial, with the same skim milk powder and samples collected from different sections of the digestive tract, as well as in a human study with from jejunal effluents. In vivo, digestion is a dynamic process influenced by peristalsis and by the gradual secretion of enzymes and juices and the dwell time of the food. To mimic these physiological mechanisms, dynamic in vitro digestion protocols are widely used. Until now, the differences of protein hydrolysis taking place during dynamic and static in vitro digestion have not been investigated.In this study, the gradual hydrolysis of the main milk proteins present in skim milk powder was digested with the dynamic DIDGI®-system using adult digestion protocol and the static harmonized INFOGEST method. Protein hydrolysis was analyzed by gel electrophoresis, peptide patterns were measured with mass spectrometry, and free amino acids with high pressure liquid chromatography. The peptide patterns at the gastric and intestinal endpoints of in vitro digestion showed a good approximation to the in vivo results from pigs. Moreover, gradual peptide generation was comparable in both in vitro digestion conditions. However, the dynamic protocol reflected the physiological situation better at the level of free amino acid release. Nonetheless, in both in vitro digestion protocols, absorption of free amino acids is not simulated, and they are therefore limited in reflecting the in vivo situation at this level.

National audience; Introduction Les nourrissons français reçoivent en majorité des préparations pour nourrissons (PPN), qui reçoivent de nombreux traitements thermiques pour assurer leur sécurité sanitaire. Ces traitements peuvent dénaturer/agréger les protéines et favoriser la réaction de Maillard, ce qui peut impacter les propriétés nutritionnelles et organoleptiques des PPN. Le but de cette étude était de développer un itinéraire technologique permettant de produire une PPN conservant ses protéines à l’état natif tout en garantissant sa qualité bactériologique grâce à la filtration membranaire, et d’en évaluer l’impact sur la digestion des protéines. Matériel et Méthodes Des PPN en poudre ont été produites à l’échelle semi-industrielle à partir de lait frais écrémé et épuré, additionné d’un concentré de protéines solubles natives, tous deux obtenus par microfiltration (1,4 et 0,1 m, respectivement). Lactose, huiles végétales, minéraux et vitamines ont été ajoutés pour obtenir une PPN complète (1er âge) au sens de la réglementation. Trois modalités ont été testées : T-, sans traitement thermique ; T+, avec pasteurisation seule du concentré (75°C, 2min) avant séchage ; T+++, combinant pasteurisation du lait (72°C, 30s) et du concentré (85°C, 2 min). Le taux de dénaturation protéique a été déterminé par Kjeldahl. Les produits de la réaction de Maillard ont été dosés par analyse de la furosine en chromatographie liquide en phase inverse et de la Nε-carboxyméthyl-lysine (CML) en spectrométrie de masse tandem (LC-MS/MS). Ces PPN ont été soumis à une digestion in vitro dynamique mimant le stade nouveau-né à terme. La cinétique de protéolyse a été évaluée par quantification des NH2 libérés et la microstructure des digestas a été suivie par granulométrie à diffraction laser et visualisée par microscopie confocale. Résultats Les PPN répondaient toutes aux exigences bactériologique et nutritionnelle, et présentaient des taux de dénaturation protéique différents (3 à 6% pour T- et T+ vs. 58% pour T+++). T+++ possédait une teneur en CML de 25% plus élevée que T- et T+, alors que des teneurs similaires en furosine étaient observées. Peu de différences ont été remarquées entre les PPN concernant le degré de protéolyse, ce qui peut masquer des différences d’hydrolyse des protéines individuelles. La microstructure des PPN était similaire avant digestion ; en revanche, les digestas provenant de T- et T+++ présentaient une microstructure différente en phase gastrique dès 80 min de digestion. Conclusion Cette étude a démontré la faisabilité de production de PPN conservant leurs protéines à l’état natif et sûres bactériologiquement. Les expérimentations de digestion in vitro ont montré que la microstructure des digestas était différente entre la PPN native et la PPN la plus traitée thermiquement, ce qui pourrait impacter les mécanismes de régulation de la digestion chez le nourrisson. La digestion des protéines individuelles sera investiguée prochainement.

International audience; In 2018, about 60% of world’s newborns received cow milk-based infant formulas (IF) instead of human milk (UNICEF). The process of manufacturing IF involves heat treatments altering the physicochemical properties of milk components, especially whey proteins (WP). The objective of the study was to investigate the impacts of thermal treatments on the denaturation of WP of IF, particularly for those mimicking the protein profile of human milk, and to characterize the heat-induced protein structures. Methods: Three model IF were produced with a caseins:WP ratio of 40:60 at 1.3% and 5.5% of total proteins, differing in the quality of WP. The kinetic of heat-induced denaturation of each WP was investigated between 67.5°C and 80°C by quantification of the residual native proteins by RP-HPLC. The heat-induced protein structures were studied by dynamic light scattering, electrophoresis and asymmetric flow field flow fractionation coupled with MALLS. Results: The quantification of native WP fractions revealed an enhanced denaturation of α-lactalbumin in presence of both β-lactoglobulin and lactoferrin, an increased denaturation of β-lactoglobulin in presence of lactoferrin, and an increased denaturation lactoferrin in presence of β-lactoglobulin. At similar rate of total WP denaturation at 67.5°C and 80°C, the protein composition of the heat-induced aggregates changed between formulas, protein concentrations and heating temperatures but disulfide bonds were the main intermolecular links. The aggregates were larger and of fractal shape (apparent fractal dimension dfapp = 2.1) in formulas at 5.5% proteins whereas they were of spherical shape (dfapp = 2.9) in formulas at 1.3% proteins. Discussion: Effets of heat treatment on the denaturation of α-lactalbumin, β-lactoglobulin and lactoferrin and the protein structure are influenced by the WP quality of IF. These results will give to industrials reliable data on the protein structures formed during the heat treatments of IF. The impact on digestibility will be subsequently investigated

National audience

The rheological properties and microstructure of dairy gels involve the connectivity between milk fat globules (MFG) and casein micelles that is affected by technological processes such as milk homogenization and heat treatment. The underlying mechanisms require further quantification of the interactions at the nanoscale level to be fully understood and controlled. In this study, we examined the adhesion of homogenized MFG to milk proteins and evaluated the role of ultra-high temperature (UHT) heat treatment and pH. The combination of physico-chemical analysis, rheology and microscopy observations at different scale levels associated to atomic force microscopy (AFM) force spectroscopy were used. AFM experiments performed at the particle scale level showed that adhesion of individual homogenized MFG to milk proteins (1) is increased upon acidification at pH 4.5: 1.4 fold for unheated samples and 3.5 fold for UHT samples, and (2) is enhanced by about 1.7 fold at pH 4.5 after UHT heat treatment of milk, from 176 pN to 296 pN, thanks to highly-reactive heat-denatured whey proteins located at the surface of MFG and caseins. The increased inter-particle adhesion forces accounted for more connected structures and stiffer UHT milk acid gels, compared to unheated-milk gels. Using a multiscale approach, this study showed that heat treatment of milk markedly affected the interactions occurring at the particle’s surface level with consequences on the bulk structural and rheological properties of acid gels. Such findings will be useful for manufacturers to modulate the texture of fermented dairy products through the tailoring of heat-induced complexation of proteins and the connectivity of homogenized MFG with the protein network. This work will also contribute in a better understanding of the impact of process-induced changes on the digestibility and metabolic fate of proteins and lipids.

Food is a complex combination of chemically diverse structures, influenced by composition and processing. The nutritional quality of food cannot be solely based on individual contents in nutrients as food structure will strongly modulate the bioaccessibility of nutrients upon digestion. In previous studies, the most common approach for measuring the extent of proteolysis and nutrient uptake was to compare the initial macroscopic features of ingested food to final concentrations in the bloodstream. However, there is limited literature to the support underlying mechanisms of the digestion of food in the stomach and enzyme diffusion in solid food substrates. In the present work, the influence of casein protein gel microstructure on enzymatic digestion was investigated for the future development of personalised foods. Casein gels with an identical composition, but differing by the coagulation mode, were characterized and submitted to simulated gastric digestion. The changes in gel structure over time were studied using confocal microscopy and small angle x-ray scattering (SAXS). The molecular interactions that occur as a gel network is formed, and subsequently broken down under digestive conditions will be discussed.

Dans l’industrie laitière et fromagère, les protéines du lactosérum ne sont pas ou que très peu valorisées dans les produits transformés. En effet il est assez aisé de se servir des protéines micellaires (caséines) pour faire des yaourts ou encore des corps gras, matière première du beurre. Il est vrai que le comportement des protéines sériques est difficilement compréhensible au premier coup d’oeil, elles sont de petites tailles et ne forment pas facilement de structures macroscopiques notables. Ceci explique qu’aucune valorisation à forte valeur ajoutée n’ai été trouvé de façon empirique. Cependant l’étude fondamentale des interactions entre ces protéines sériques est très prometteuse, nous savons qu’elles ont tendance à former des coacervats dont l’exploitation est indéniablement vouée à de nombreuses applications (encapsulation, protection de molécules d’intérêts, …). L’équipe ISFPL du STLO travaille depuis de nombreuses années sur l’étude de ces interactions. Ils les ont déjà étudiés par plusieurs méthodes physico-chimiques : titration calorimétrique isotherme (ITC), fluorescence, dialyse à l’équilibre, … La majorité de ces méthodes d’analyses sont appliquées sur un très grand nombre de molécules et résultent donc d’un comportement de milieu. Dans le cadre de mon stage, nous nous intéresserons uniquement à l’interaction entre deux protéines individuelles. Au sein du STLO, nous avons la chance de posséder un microscope à force atomique qui permet de réaliser des spectres de forces d’interactions entre deux molécules. L’objectif de mon stage est d’obtenir et d’interpréter des spectres de forces résultant des interactions entre protéines sériques., In dairy and cheese industry, whey proteins are not too much used. In fact, understand their chemical comportement isn’t as easy as micellar proteins (yoghurt base) or as fatty substance (butter base) because of their little size and their hard coagulation. However, the fundamental study of whey protein’s interactions can be really interresting. They make coacervates with lot of applications. The ISFPL team is working on this subject since many years, they already studied it by isotherm titration calorimetry, fluorescence, nuclar magnetic resonance 2D, ... Most of them measure moleculs in group and don’t give us informations about just two protein’s interaction. During my practice I will learn how to use an atomic force microscop, in order to establish force spectres. The results would help us to understand better whey protein’s interactions.

Dry heating (DH) at 100 °C of a whey protein isolate powder conditioned at pH 9.5 leads to the formation of stable, large and porous whey protein microparticles (WPM), resulting from crosslinking of proteins in the powder. The Maillard reaction (MR) is probably involved in the formation of these WPM. To confirm the role played by the MR and understand the protein changes occurring during particle formation, we analyzed the properties of the powders (such as their water activity (aw) and color) and characterized the WPM formed after 3 and 36 h of DH in powders with increasing amounts of added lactose. Adding lactose and prolonging the time of DH increased the extent of protein denaturation and the rate of formation of WPM and the yield of conversion of powder into WPM, but led to decreases in their water content. Addition of lactose also altered the kinetics of the MR, the increase in aw and the formation of WPM. The MR, as followed by powder browning, was complete within 3 h of DH irrespective of the amount of added lactose. In contrast, the increase in aw, thought to be due to progress of the MR, continued to increase from 3 to 36 h of DH. Formation of WPM went on increasing from 3 to 36 h of DH at low to medium levels of lactose, but was complete by 3 h of DH at high levels of lactose. In conclusion, the formation of WPM was not simply related to the MR and needs further investigation.

Beyond nutrient composition matrix plays an important role on food health potential, notably acting on the kinetics of nutrient release, and finally on their bioavailability. This is particularly true for dairy products that present both solid (cheeses), semi-solid (yogurts) and liquid (milks) matrices. The main objective of this narrative review has been to synthesize available data in relation with the impact of physical structure of main dairy matrices on nutrient bio-accessibility, bioavailability and metabolic effects, in vitro, in animals and in humans. Focus has been made on dairy nutrients the most studied, i.e., proteins, lipids and calcium. Data collected show different kinetics of bioavailability of amino acids, fatty acids and calcium according to the physicochemical parameters of these matrices, including compactness, hardness, elasticity, protein/lipid ratio, P/Ca ratio, effect of ferments, size of fat globules, and possibly other qualitative parameters yet to be discovered. This could be of great interest for the development of innovative dairy products for older populations, sometimes in protein denutrition or with poor dentition, involving the development of dairy matrices with optimized metabolic effects by playing on gastric retention time and thus on the kinetics of release of the amino acids within bloodstream.

Dry heating (DH) at 100 °C for 36 h of a whey protein isolate powder conditioned at pH 9.5 leads to the formation of stable, large and porous whey protein microparticles (PMs), resulting from the crosslinking of proteins inside the powder. These PMs could be used as high-viscosity food ingredients. Casein, present as a contaminant in whey protein powders, has been shown to become incorporated into the PMs. In this study, we investigated the effect of adding increasing amounts of sodium caseinate to whey protein powders on the formation of PMs during DH at 100 °C for 36 h. In addition, we studied PM formation during DH of a micellar casein-enriched milk protein powder (Casmic). The browning index of the dry-heated powders, and the size and water content of the microparticles were also characterized. We confirmed that sodium caseinate was incorporated into the PMs. The highest PM D[4,3] values (270 μm) were observed for powders with around 40% caseinate. Powders without added caseinate displayed D[4,3] values of 150 μm. The yield of conversion of proteins into PMs increased from 0.6 to 0.8 g/g with caseinate addition, whereas the amount of water entrapped in the PMs decreased from around 30 to 20 g/g. PMs were also formed by DH of the Casmic powder, but these particles were smaller, with sizes of around 80 μm. In conclusion, our study shows that the process of DH at pH 9.5 could be applied to all milk proteins to obtain PMs with functional properties that could be used in the food industry.

International audience; Infant formulas (IFs) can be defined as substitutes for human milk, which are mostly based on cow milk proteins. For sustainability reasons, alternative to animal proteins in food have to be considered. Plant proteins offer interesting nutritional and functional benefits for the development of innovative IFs. However, the behaviour of these proteins during processing and storage must ensure the physical stability and ability to reconstitution of IF powders, and that needs to be tested. This work aimed to study how a partial substitution of dairy proteins by plant proteins may influence the functional properties of 1st age IFs. Three IFs were developed at a semiindustrial scale using two different processing routes. The IFs composition was identical, except that 50% of the proteins were whey proteins in the “reference IF” (RIF), and pea or faba bean proteins in the “plant IFs” (PIF and FIF, respectively). After reconstitution, the three IFs result in similar stable emulsions with equivalent free fat release. In comparison to RIF, PIF and FIF were difficult to disperse, conducting to remaining insoluble particles. Thus, the protein source greatly influences IFs properties, and process parameters need to be adapted for each formulation to meet IFs quality criteria.

Introduction et but de l’étude Les protéines généralement utilisées dans les formules infantiles proviennent du lait de vache et sont majoritairement des protéines de lactosérum afin de se rapprocher de la composition du lait maternel. Cependant, certaines protéines végétales offrent également de bonnes propriétés nutritionnelles et pourraient constituer d’intéressants substituts dans un contexte de compétition d’usage des protéines de lactosérum. Le but de cette étude était de comprendre l’impact de la substitution des protéines de lactosérum par des protéines végétales sur le degré d’hydrolyse et sur la bioaccessibilité des acides aminés. Matériels et Méthode Quatre protéines végétales ont été sélectionnées (PV1 à PV4) en raison de leur composition en acides aminés essentiels adaptée aux besoins du nourrisson. Elles ont été utilisées pour développer des formules infantiles de composition identique à la référence (R), à ceci près que les protéines de lactosérum étaient remplacées par une des quatre protéines végétales. Les cinq formules infantiles ont ensuite été soumises à un modèle de digestion in vitro gastro-intestinale en conditions statiques (Ménard et al. 2018). Le degré d’hydrolyse des protéines était mesuré par la méthode OPA (O-phtaldialdehyde). Les acides aminés libres ont été quantifiés par chromatographie échangeuse de cations. Résultats Tableau 1. Teneur en acides aminés libres et degré d’hydrolyse en point final de digestion de cinq formules infantiles Formules infantiles Thréonine (%) Isoleucine (%) Leucine(%) Lysine (%) AAS (%) AAA (%) Degré d’hydrolyse (%) R 12,6±0,7 19,1±0,6 47,9±2,9 71,1±3,2 23,6 ±2,6 79,6 ±3,7 53,0±6,1 PV1 5,8±0,1*** 1,9±0,2*** 12,1±0,7*** 16,9±0,9*** 9,1 ±1,2*** 21,1±0,9*** 29,0±4,1*** PV2 15,7±1,2* 20,9±0,7 51,4±1,6 72,6±1,1 22,0 ±2,3 74,8 ±5,2 65,6±1,2 PV3 14,9±4,0 17,2±1,1 49,7±4,9 73,1±4,4 24,1 ±4,8 75,5 ±5,3 45,4±4,4 PV4 11,8±2,3 16,4±2,8* 45,6±5,1* 68,7±6,6 25,8 ±6,8 71,1 ±8,2 40,2±2,7* Les différences entre les valeurs moyennes des formules PV1 à 4 par rapport à R sont considérées significatives lorsque p < 0.05 (t-Test : *0,05 ; **0,01 ; ***0,001). AAS (acides aminés soufrés) ; AAA (acides aminés aromatiques) Les formules PV2 et PV3 libèrent autant d’acides aminés que la référence (R) et présentent un degré d’hydrolyse équivalent (p>0.05). En revanche, la formule PV4 se distingue de R par une moindre libération d’isoleucine et de leucine et présente également un degré d’hydrolyse plus faible. Enfin, la formule PV1 conduit à des valeurs très inférieures à celles obtenues avec R pour toutes les variables étudiées. Conclusion Il est établi que la cinétique de digestion des protéines dans une matrice et par conséquence, la libération des acides aminés et leur absorption peuvent être influencées notamment par sa composition en protéines (Le Feunteun et al., 2014). De manière concordante, nous montrons ici que la nature des protéines des formules infantiles aurait un effet sur la protéolyse. Ainsi, parmi les quatre formules « végétalisées » testées, seules PV2 et PV3 sembleraient être nutritionnellement comparables à la référence. Des digestions in vitro en conditions dynamiques (plus proches des conditions physiologiques) seront réalisées prochainement afin de compléter ces résultats quant à la qualité nutritionnelle de ces formules infantiles innovantes.

Les recommandations de l’OMS pour l’alimentation du nourrisson et du jeune enfant sont un allaitement exclusif jusqu’à l’âge de 6 mois, puis un allaitement partiel jusqu’à l’âge de 2 ans. Le taux et la durée de l’allaitement maternel en France restent cependant plus bas que les autres pays d’Europe. En 2017 en France, seulement 38% des enfants sont allaités jusqu’à l’âge de 6 mois (WHO/UNICEF, 2017). Lorsque l’allaitement maternel n’est pas possible ou non souhaité pour diverses raisons culturelles, économiques ou personnelles, les formules infantiles restent l’alternative la plus adaptée à la nutrition du nourrisson si leurs compositions respectent les besoins nutritionnels du jeune enfant. Ces formules doivent participer au développement optimale du nourrisson tout en respectant ses contraintes physiologiques, liées notamment à une immaturité digestive et rénale. Les formules infantiles doivent donc se rapprocher le plus possible du lait maternel. Leur composition nutritionnelle est globalement très proche, mais la composition fine et la structure diffèrent de par l’organisation des gouttelettes lipidiques et de la nature des protéines. En terme de digestion, étant donné l’immaturité du tube digestif à la naissance, des conditions de spécificités digestives adaptées aux nourrissons sont à prendre en considération afin de mettre en place des modèles de digestion in vitro pertinents (Bourlieu et al., 2014). Ces modèles sont utilisés aujourd’hui pour des raisons éthique, économique et de praticité pour les études de digestibilité. L’amélioration des formules infantiles, via la recherche et l’innovation, permet de répondre aux besoins nutritionnels et aux enjeux économiques. Des formules infantiles ont été développées pour répondre aux difficultés telles que les allergies, les intolérances, les problèmes de régurgitation, les troubles digestifs ou encore une production plus respectueuse de l’environnement. L’incorporation de probiotiques pour reproduire une partie du microbiote du lait maternel fait également partie de ces nouvelles générations de formules infantiles, mais également des formules infantiles de structures lipidiques et de composition en protéines plus proche du lait maternel ont été mises au point. De nouvelles perspectives visent à rechercher des protéines alternatives aux protéines de lactosérum du lait de vache via l’introduction de nouvelles sources de protéines d’origines végétales. L’objectif est de rechercher et d’étudier l’apport de nouvelles sources protéiques et d’aller jusqu’à la faisabilité technologique et la mesure de digestibilité en vue de s’assurer de la valeur nutritionnelle de ces nouvelles sources protéiques.

National audience; Whey protein (WP) have been extensively studied in the past and are commonly used for their functional and nutritional properties in food products. Recently, a particular interest was drawn in designing whey protein aggregates by adjusting physicochemical (pH, salt and protein concentration) and processing conditions in order to obtain new functional properties, depending on aggregate structure and size obtained (compact sphere, fibril or strand like). Namely, WP aggregates obtained at pH 7 and 80°C present a fractal structure that could be interesting for increasing dairy products texture. In this view, the aim of this study was to assess the influence of pilot scale continuous process conditions (temperature rise kinetics, outlet heating temperature, Reynolds number value) on the structure, size and properties of experimentally elaborated WP aggregates. Aggregates were obtained from the heat treatment of a WP solution (5% w/w), with a pH adjusted at 7 and a salt concentration of 0,01 M. The heat treatment pilot included a feed tank, a preheating exchanger to increase the temperature up to 50°C or 55°C, a tubular heat exchanger to achieve the target outlet heating temperature (80°C or 85°C) and finally a water-bath for a static holding. Moreover, it was possible to vary in an independent way the temperature rise kinetics and the flow regime depending on the mass flow rate and the tubular heat exchanger inner diameter chosen (6 to 22,5 mm). WP aggregates were separated and analysed by Asymetrical Flow Field-Flow Fractionnation coupled with Multi-Angle Light Scattering and Differential Refractometer (AF4-MALS-DRi) and by Transmission Electron Microscopy (TEM). After 80°C heat treatment, three populations of aggregates were observed in the solution : non-aggregated proteins, and 2 types of WP aggregates with an average radius of gyration (Rg) of 10 nm for the first, and 60 nm to 100nm for the second one. It was shown that both temperature rise kinetics and flow regime affected mean size of this latter : namely decreasing Reynolds number from 10000 to 2000 induced an increase of the aggregates size. Moreover, increasing temperature led to the formation of a fourth population of larger aggregates (Rg ≥200nm). These results show that process parameters are efficient levers to modulate the size and structural properties of fractal aggregates in a semi-industrial scale and should be carefully assessed in order to monitor their functional properties.