Résumé
Human milk is the ideal food for the newborn infant (WHO, 2015). Donor human milk is pasteurized for sanitary reasons, which can affect the gastric hydrolysis of intact proteins (De Oliveira et al., 2017). This study aimed to evaluate whether this also affects the kinetics of peptide release during gastric digestion in the preterm infant. Twelve hospitalized preterm infants fed by a nasogastric tube received daily the own mother’s milk as raw (RHM) or pasteurized (PHM), over 6 days (NCT02112331). Gastric content was withdrawn before the meal and either 35, 60 or 90 min after the feeding start. The postprandial sampling was conducted in duplicate on two different meals. Samples were analysed semi-quantitatively by tandem mass spectrometry coupled to nano-chromatography. After data transformation, a mixed linear model was applied to each peptide with the main factors being postprandial time, meal and their interaction. All interaction coefficients were subjected to an ascending hierarchical classification. Five clusters were identified over 1579 peptides, originating at 58% from β-casein and at 10% from lactoferrin. Clusters were significantly characterized by the parent protein and amino acid cut-off. Over three clusters (1, 3, 4; 32% of the peptides), peptide abundancy was significantly higher for PHM even before digestion, with the parent protein being represented more by β-casein and less by lactoferrin than in overall. On cluster 5 (23%), peptide abundancy was slightly higher for PHM. On cluster 2 (44% of the peptides), there was no difference of abundancy between RHM and PHM over the digestion course; the parent protein was more represented by lactoferrin than in overall. Pasteurization impacted the kinetics of peptide release for 56% of the peptides, which could partly result from the activation of the milk proteolytic system of plasmin (Deglaire et al., 2016). These modifications could impact the peptide bioactivities and/or amino acid absorption kinetics.